1. Introduction

• Can artificial intelligence aid in the identification and management of symptoms associated with anxiety and depression?
• Which artificial intelligence techniques are currently employed for symptom management of anxiety and depression, and how is their efficacy evaluated?
• What potential benefits and dangers are arising from the utilization of these technologies by individuals and healthcare practitioners?

2. Background: The Role of Informatics and AI in Mental Health

2. Methods

3. Results of Overview

3.3. AI Methodologies and Tools Utilized for Anxiety and Depression Management

AI tools for anxiety and depression encompass many technologies specifically developed to evaluate, control, and address mental diseases. Examples of such technologies include chatbots and virtual assistants, mood tracking applications, cognitive behavioral therapy (CBT) apps, and virtual reality (VR) therapies [51]. AI-driven chatbots and virtual assistants utilize user engagement to provide emotional support, education, and coping mechanisms for managing symptoms associated with anxiety and depression [14]. These technologies employ natural language processing (NLP) algorithms to comprehend user data and deliver suitable replies. Conversely, these applications track users’ emotional condition, symptoms, factors, and actions associated with anxiety and depression. Consequently, these insights offer valuable information on recurring patterns and long-term trends, enabling users to make well-informed choices regarding their mental well-being [4]. However, AI-driven CBT programs offer scientifically supported therapy approaches for anxiety and depression, including cognitive restructuring, behavioral activation, and relaxation strategies. These applications have the ability to include interactive activities, modules that may be completed independently, and customized feedback depending on the user’s input and tracking of their progress [79]. Moreover, VR-based therapies provide lifelike surroundings and scenarios to expose patients to stressful events or teach them relaxation and coping skills in a controlled setting. AI algorithms have the ability to tailor the virtual reality experience in real-time by analyzing user emotions and physiological data. This enhances the effectiveness of treatments and allows for a more personalized experience [9].

AI-powered algorithms evaluate physiological data obtained from mobile devices, including heart rate variability, sleep patterns, activity levels, and speech patterns, to detect small changes that may indicate symptoms of anxiety and sadness [58]. Digital biomarkers provide precise and unbiased assessments of an individual’s mental health condition and their reaction to therapy, enabling prompt intervention and tailored healthcare. By utilizing AI ML tools in these several domains, mental health practitioners may enhance the precision of assessments, the efficacy of treatments, and the involvement of patients in the management of anxiety and depression. The following table compiles technologies that employ artificial intelligence to assist users in identifying and handling symptoms associated with anxiety and depression. Later on, a subset of them will undergo more comprehensive analysis in order to comprehend their purpose and influence.

Figure 1. Indicative illustration of mental health chatbots, like WYSA, using generative artificial intelligence.

Figure 1. Indicative illustration of mental health chatbots, like WYSA, using generative artificial intelligence.

Table 1. This is a table. Compiles technologies that employ artificial intelligence to assist users in identifying and handling symptoms associated with anxiety and depression.

Table 1. This is a table. Compiles technologies that employ artificial intelligence to assist users in identifying and handling symptoms associated with anxiety and depression.

CHATBOTS	Woebot	Woebot is an AI-driven chatbot that assists individuals in managing their mental well-being through the application of cognitive-behavioral therapy approaches. The application offers mood monitoring, daily assessments, and tailored dialogues. As the current leading application, Woebot is likely to continue improving and expanding [28].
	Wysa	Wysa is an AI chatbot that utilizes cognitive-behavioral therapy techniques to offer support for mental health. Its features include guided meditation, mood tracking, and tailored conversations. Wysa is widely used by individuals seeking mental health support and has garnered positive reviews [39].
	Youper	Youper exemplifies the integration of artificial intelligence (AI) into the realm of mental health care. As a health technology business, the firm aims to ensure that mental health treatment is accessible and affordable for all individuals. Youper AI assistant interacts with individuals in meaningful dialogues to assess their psychological condition and offers tailored remedies based on the gathered information. This groundbreaking approach guarantees that people receive customized assistance, highlighting the transformative capacity of AI in revolutionizing mental healthcare [55].
	Replika	Replika is an artificial intelligence chatbot that constructs a virtual representation of users by analyzing their personality traits. Its ultimate goal is to help users cope with stress and improve their mental health. It is suitable for individuals who desire to engage in profound and contemplative conversations with a companion. It offers users a safe space for self-reflection and emotional assistance, fostering meaningful connections and understanding. Replika has emerged as a powerful tool for those seeking alternative methods to safeguard and improve their mental well-being [66].
	Lifeline Ally	A friendly chatbot that focuses on preventing and supporting depression [101].
	Humorous Psychological Companion	The Humorous Psychological Companion is a distinctive combination of a digital assistant and conversational partner, particularly created to aid individuals who are dealing with depression or related emotional conditions. It serves not just as an AI chatbot, but also as a sympathetic, clever, and empathetic friend. The primary objective of this platform is to aid, motivation, and a cheerful conversation to mitigate emotions of sorrow or isolation [106].
	Elomia	Elomia is an AI-powered virtual therapist who has been trained in thousands of consultations, offering quality advice and support. Users can discuss their problems, ask questions, and get recommendations for mental health exercises. According to research, Elomia can help with anxiety, depression, low self-esteem, loneliness, relationship issues, burnout, and sleep problems. Whenever users need someone to talk to or seek guidance, Elomia listens, helps identify concerns and suggest solutions, helping to regain confidence and acceptance of emotions [70].
	Tess	Tess, on the other hand, is a mental health chatbot that provides treatment and support to people experiencing symptoms of depression and anxiety [29].
	Meru Health	Meru Health is an artificial intelligence app that offers treatments for depression, anxiety, and stress. It provides personalized treatment regimens that meet the specific needs of each user, as well as online counseling and guidance [107].
APPLICATIONS	Ginger	Ginger is an on-demand mental health platform that provides counseling, support, and advice. It offers chat therapy, individualized care plans, and video sessions with licensed therapists [104].
	Headspace	Headspace, a mindfulness and meditation software, makes individualized recommendations using artificial intelligence algorithms. It offers daily reminders, sleep sounds, and guided meditations to help users stick to their mental health goals [105].
	Breathhh	Breathhh is a Chrome plugin that uses artificial intelligence to deliver mental health workouts tailored to an individual’s online activities and behavior. Through the surveillance and examination of user interactions, Breathhh is able to ascertain the optimal moments to introduce stress alleviation methods and tactics. This innovative method blends artificial cognitive intelligence with the tangible assistance of mental health, facilitating a smooth and effortless incorporation of self-care into users’ everyday schedules [102].
	Sanvello	Sanvello is an app for mental wellbeing that tracks users’ moods and helps them understand their situation. It offers individualized mental health care. It also provides basic principles through the use of ambient noise. Through a community where users may engage in conversations with others facing similar problems, it also offers peer-to-peer help [15].
	MindDoc	MindDoc provides a number of tools to help different facets of mental health, such as enhancing positive coping mechanisms, tracking your mood, or monitoring your general well-being. Her area of expertise is understanding mental health issues like sleeplessness, eating disorders, anxiety, and depression. It is simple to obtain helpful materials, exercises, and customized recommendations thanks to the user-friendly layout [109].
	MoodMission	This program is designed to aid users in overcoming depression and anxiety by implementing evidence-based coping strategies. MoodMission facilitates personal growth and self-empowerment by suggesting tailored assignments that correspond to the emotions and experiences expressed by the user. By successfully completing missions, you will improve your understanding of your mental health and discover new techniques to efficiently manage difficult tasks [10].
	Ladder	Ladder is a health application that uses artificial intelligence to assist users in comprehending the correlation between their behaviors, emotions, and their general state of well-being. The software’s most compelling attribute lies in its origin and purpose, since it was developed only by and for individuals belonging to ethnic minority groups. This multifunctional device incorporates an exercise tracker to facilitate the development of beneficial routines, while its cognitive journal promotes emotional mindfulness and introspection [103].
	Kintsugi	Kintsugi employs an innovative method in the field of mental health treatment by utilizing cutting-edge voice biomarkers in speech analysis to promptly detect, rank, and tackle mental health concerns as they arise. This API-centric platform readily interfaces with contact centers, telemedicine systems, and remote patient monitoring apps, enabling enhanced accessibility to appropriate treatment as and when required. Through the identification of small alterations in speech patterns and vocal indicators, it is capable of precisely evaluating an individual’s psychological condition and guiding them towards the most suitable resources and assistance [106].
	Calm	Calm has established itself as a highly regarded application for mental well-being, particularly for individuals grappling with anxiety and despair. This application offers a range of powerful resources, including sleep tales, meditation, and other methods, to assist users in managing their mental health difficulties. Users have the opportunity to investigate a range of characteristics that cater to their own requirements in order to enhance their overall state of being [100].
	Rootd	Rooted is a novel application that provides valuable assistance to individuals experiencing panic attacks and anxiety. The Rootr feature of the app offers individuals a convenient way to manage stress and find immediate relief. Furthermore, Rootd provides a range of strategies, including exercises, routines, and healthy diets, to effectively promote mental health improvement. By implementing these strategies and utilizing these resources, users can gradually improve their capacity to manage stress and achieve a more harmonious lifestyle [116].
	MindShift	MindShift is a very helpful mental health tool that helps users properly control their anxiety by applying tried-and-true methods from Cognitive Behavioral Therapy (CBT). In addition to encouraging the development of constructive thinking patterns and the adoption of preventative actions to successfully manage stress-related difficulties, the program gives users the chance to practice mindfulness and relaxation. With MindShift’s exquisitely simple interface and numerous adaptable options, users can effortlessly develop relaxing routines and enhance their mental health in general [69].
	Happify	Happify provides customers with interactive activities and games specifically intended to effectively alleviate stress and counteract negative thinking. The software offers customized analytics derived from an individual’s mental health data, guaranteeing tailored assistance that caters to their distinct requirements. Users may quickly access these activities and games at any time, allowing for seamless integration into their everyday routine. Happify enhances happiness and mental well-being, thereby enhancing an individual’s entire quality of life through its creative method [123].
	Deepwander	Deepwander is an AI-driven platform or tool created to enhance self-reflection and individual development. It involves participants in interactive discussions intended at investigating their internal realm and directing them towards beneficial transformation. Deepwander utilizes a range of psychological strategies, including Cognitive Behavioral strategies, Motivational Interviewing, Narrative Therapy, and Guided Visualization, to assist users in altering negative thinking patterns, discovering motivation, restructuring their life narratives, and clarifying their objectives. In general, Deepwander appears to offer a systematic method for introspection and individual growth [105].
		MindwellAI is a mental health app designed to help users overcome stress. It combines science-based tools, AI-powered counseling, and a virtual self-care partner named Joy [111].
	Space of Mind	Space of Mind is an online trauma therapy and support group designed to provide affordable PTSD treatment. This AI-powered platform offers a therapeutic space facilitated by a therapist where individuals can participate in anonymous group sessions to address their traumatic experiences and work towards transformative therapy [120].
	Moodpath	Moodpath is a mental health assessment tool that helps individuals understand their mental well-being through science-based questions and assessments. It identifies early signs of depression, anxiety, and burnout, offering personalized diagnoses and access to support resources. Users get insights into their mental health, trends, and tips, along with tailored action plans for long-term management. In addition, Moodpath offers meditations, self-care tips, and personalized plans, making it a valuable resource for managing mental health [115].
	MoodKit	MoodKit is a mobile app designed to help individuals manage their mood and create healthy emotional habits. Developed by psychologists, MoodKit offers a variety of tools and resources based on the principles of cognitive behavioral therapy (CBT), a widely recognized and effective form of treatment for mood disorders [10].
WEARABLES DEVICES	Samsung Health	Samsung Health is an all-encompassing health and wellness application created by Samsung Electronics. The purpose of this application is to assist users in tracking and controlling several elements of their well-being and physical condition, such as exercise, nutrition, sleep, stress levels, and more. The application provides functionalities such as step monitoring, exercise monitoring, calorie calculation, sleep monitoring, and tools for managing stress [127].
	MoodTools	MoodTools is a mobile app designed to provide support and resources for people experiencing depression. It offers various tools and features that help users effectively manage their mood and mental health [113].
	BioBase	BioBase is a wearable device designed to monitor and analyze physiological signals in real time, providing information about a person’s stress levels, energy, and overall well-being. It uses biometric sensors to monitor parameters such as heart rate variability (HRV), skin conductivity and temperature, which are indicators of stress and arousal levels [45].
	Spire Health Tag	The Spire Health Tag is a portable health device that monitors various aspects of well-being throughout the day. Small labels attach to your clothes and track your activity levels, breathing patterns, stress levels, and sleep quality. Using advanced sensors, Spire Health Tag provides real-time feedback and information about your daily habits and how they affect your overall health [128].
	Moodmetric	Moodmetric is a wearable device designed to monitor and manage stress levels in real time. It measures electrodermal activity (EDA), the electrical activity of the skin caused by the activation of sweat glands, to provide information about stress and emotional arousal [122].
	Fisher Wallace Stimulator	The Fisher Wallace Stimulator is a small device designed particularly to aid in the treatment of anxiety, depression, and insomnia. The procedure is applying gentle electrical stimulation to the brain using small electrodes placed on the forehead. The aim of this stimulation is to regulate the levels of neurotransmitters, such as serotonin and cortisol, which are associated with mood and anxiety. Multiple users have claimed positive results, such as reduced anxiety and improved mood, however individual outcomes may vary [125].

Figure 2. Distribution of mental health Chatbots.

Figure 2. Distribution of mental health Chatbots.

Figure 3. Distribution of mental health Applications.

Figure 3. Distribution of mental health Applications.

Figure 4. Distribution of mental health Wearable Devices.

Figure 4. Distribution of mental health Wearable Devices.

3.3.1. Mobile Apps

The ratio of psychiatrists per inhabitant of the world is less than satisfactory. However, over 50% of the world’s population now owns a smartphone [51]. Based on this data, many mobile applications have been designed in recent years, which offer psychological support, breathing, and relaxation exercises, while collecting useful data on the health and well-being of users. Moreover, wearables, platforms, and mobile applications have integrated artificial intelligence technologies, hence creating new opportunities for diagnosing and treating symptoms related to these conditions. Mobile applications (apps) designed for mental health purposes have become readily available tools that provide regular monitoring of mental well-being and assistance [36]. These applications provide a convenient and often anonymous substitute for conventional medicinal methods.

Mobile applications that use artificial intelligence for mental health rely on many data sources to provide personalized advice and assistance to their users. The inputs may consist of data provided by users via stress and mood questionnaires, as well as data obtained from wearables or mobile sensors that capture information such as heart rate, sleep habits, and activity levels. Applications that gather and analyze this data have the ability to detect trends and symptoms that signify a higher likelihood of experiencing anxiety or depression. Additionally, they may provide timely alerts or recommendations for intervention [58]. These applications have diverse applications and users, serving various purposes. They are utilized by individuals seeking self-help, medical professionals as supplementary tools in treatment, and researchers studying their efficacy [35]. Users are granted access to a diverse range of information and services, including regular monitoring of their mental well-being, recommendations for self-care activities, availability of crisis management tools, and participation in online support communities [16]. Medical providers may use apps to remotely monitor their patients’ development and offer them more personalized care [16]. Some examples of these applications are Sanvello, MindDoc, MoodMission, Ladder, Kintsugi, Calm, Rooted, and MindShift. Each of these programs provides users with a distinct range of features and methods to help them with their daily routines.

Sanvello offers individuals personalized guidance and strategies to address feelings of melancholy and anxiety via the use of mindfulness, relaxation, and cognitive behavioral therapy (CBT). Users have the capability to regulate their mood and anxiety levels, enabling them to track their progress and get prompt feedback and suggestions [15]. MindDoc’s main goal is to provide consumers thorough evaluations of their mental well-being via daily quizzes and progress reports [108]. Utilizing this application may assist users in gaining a deeper comprehension of the variables that impact their mental well-being and identifying recurring trends in their emotional state. Consequently, this facilitates the development of effective coping mechanisms for disorders such as anxiety and depression [108].

Conversely, MoodMission recommends specific actions that users can take to alleviate their symptoms of anxiety and depression. Each “mission” is customized to the user’s current emotions and requirements, thereby facilitating the immediate alleviation of symptoms and the cultivation of self-regulation abilities [110]. Ladder provides a structured program that assists users in establishing and achieving self-improvement objectives, while also offering guidance and support. This method assists users in integrating minor, daily modifications into their regimens, thereby enhancing their overall mental health [113]. Kintsugi employs machine learning technologies to analyze audio data of users’ voices and detect indicators of anxiety and depression [105]. This novel method of monitoring enables the precise identification and monitoring of symptoms, thereby enabling users to identify alterations in their mental health.

Calm emphasizes the promotion of stress reduction and relaxation through guided breathing exercises, sleep sounds, and meditations. Users who are seeking tools to manage daily tension and enhance their sleep quality will find this application to be especially beneficial [100]. Rootd’s panic attack management system is comprehensive, offering users immediate strategies for managing panic attacks and training in long-term anxiety management strategies [115]. Cognitive behavioral therapy (CBT) and other psychological support techniques are employed by MindShift to assist users in altering anxiety-related thoughts and behaviors. Users have access to resources that can assist them in the management of anxiety and the cultivation of healthy thought processes [98].

One of these apps is MoodKit, designed to help users experiencing depression. Based on cognitive behavioral therapy, it offers a variety of tools to guide the sufferer by challenging problematic thoughts and providing space to monitor symptoms. Finally, there is a calendar for users to record important events or information for easy access in the future. Another app to regulate stress levels is Samsung Health, which monitors heart rate, among other things. When the user is stressed, they can select the guided deep breathing mode. In addition to these, there is “MoodTools”, an application designed to help people suffering from depression. In essence, the application provides access to useful videos that can have a positive impact on mood and behavior. MoodTools uses the principles of cognitive behavioral therapy to allow users to record and study their thought patterns. Very importantly, the application provides the ability to create a “suicide safety plan” to manage suicidal thoughts and crises [111,114,123].

The ubiquity of these applications in daily life is amazing. Users may integrate the applications into their daily regimen by regularly engaging with the app in the morning, tracking their emotional state and stress levels throughout the day, and participating in guided relaxation or meditation techniques in the evening. Thus, these applications integrate into everyday mental health treatment, offering continuous assistance and augmenting self-awareness [56]. The outcomes of using these applications are often favorable. Studies have shown that using artificial intelligence (AI) in the field of mental health may effectively alleviate symptoms of sadness and anxiety, boost the quality of sleep, and promote an overall feeling of well-being. Users have reported experiencing more support and a greater sense of agency in managing their mental well-being. Nevertheless, there are obstacles to overcome, such as the need to safeguard data confidentiality and prevent excessive dependence on the applications [89]. AI applications for mental health have a substantial influence on the everyday lives of its users. Users may enhance their quality of life by efficiently managing anxiety and depression via simple access to mental health management resources and the integration of self-care practices into their daily routine.

3.3.2. Chatbots

Internet-based cognitive behavioral treatment (iCBT) has been available since the 1990s, although it is known for having a poor rate of adherence [80]. CBT chatbots, designed to emulate regular chat interactions, can enhance emotional connection and deliver supplementary benefits. One such platform is “Woebot”, which offers direct assistance to patients and members through chat, along with extensive tools and information to optimize their treatment, improve the general health of the population, and increase the efficiency of healthcare practitioners. Extensive study supports it, and it has been utilized by 1.5 million individuals. This platform is specifically designed to target those who may be on the periphery of conventional healthcare, aiding that goes beyond the confines of regular clinical environments [28].

Youper is a virtual therapist that uses artificial intelligence to provide cognitive behavioral therapy (CBT) using a smartphone app. Youper enables users to engage in Cognitive Behavioral Therapy (CBT) sessions directly on their mobile devices, allowing them to specifically address and manage symptoms associated with different mental health disorders, such as anxiety or depression. The application utilizes artificial intelligence to tailor the therapy process, offering personalized exercises, strategies, and concepts that are specifically designed to meet the particular requirements and objectives of the user [55,121].

Tess, however, is a chatbot focused on mental health, namely providing therapy and support to individuals who are exhibiting signs of despair and anxiety. “Tess” is an artificial intelligence program that mimics human behavior and offers techniques for dealing with emotions shown by the user [29]. Research data indicate that Tess has demonstrated a 28% reduction in depression and an 18% reduction in anxiety [120]. The most effective utilization of such a chatbot remains uncertain. Virtual therapists who aid those with anxiety and depression are currently accessible in the market. Extensive study has been conducted to assess their efficacy [89]. In addition to providing tailored therapeutic benefits, they also enhance accessibility by circumventing some stigmas associated with mental illness.

3.3.3. Wearables and Biosignals

Technology has become a powerful ally in dealing with anxiety and managing the negative emotions associated with depression. Wearables, also known as portable devices, are an example of wearable computational models used by people to monitor their health and well-being [52]. These devices are worn on various parts of the body, such as the wrist, hand, and neck, and have sensors that continuously measure physiological signals, such as heartbeat, temperature, and galvanic skin response [20].

The main features of wearables are the ability to connect to the internet for transmitting, recording, or analyzing data, as well as the ability to connect to other electronic devices to expand their functions. Today, wearables are widely used for health management, as they have intelligent detection and communication capabilities [82]. In addition, research focuses on improving the detection capabilities of wearables using machine learning algorithms, as well as designing interventions such as deep breathing guidance. Among the wearable devices is BioBase, which measures heart rate variability and other physiological parameters to assess stress, mood, and emotional well-being [67]. It provides real-time feedback and personalized interventions to help users manage symptoms of depression and anxiety. On the other hand, the “Spire Health Tag” helps monitor the symptoms of depression. By monitoring changes in breathing and activity levels, it can detect periods of increased stress or agitation, which may be associated with depressive episodes. Finally, “Moodmetric” is a portable ring that measures electrodermal activity to assess stress and emotional arousal. It provides information about everyday stressors and helps users identify triggers for depressive symptoms, allowing them to take preventative measures to manage their mental health [116,118,124].

Smartwatches have also emerged as popular wearable devices, known for their stress reduction capabilities [36]. These devices usually monitor the user’s heart rate throughout the day, providing information on stress levels, since stress and anxiety are often associated with increased heart rate. Some smartwatches even provide built-in relaxation or breathing exercises to further aid users in stress management. Alongside smartwatches, Fisher Wallace Labs has released portable brain stimulation computing, which is reported to help reduce depressive symptoms. Their innovative devices aim to treat depression, anxiety, and other neuro-psychiatric and cognitive disorders, highlighting the potential of wearable computing in addressing mental health problems [121].

3.6. LLMs in the Treatment of Anxiety and Depression

Sophisticated technologies that have the potential to significantly enhance the treatment of mental health issues, including anxiety and depression, have been developed as a result of advancements in artificial intelligence. This chapter review investigates the potential advantages and disadvantages of large language models (LLMs) in the treatment of anxiety and depression in the field of psychiatry and psychology. Large language models (LLMs) have been extensively employed in the field of mental health care, with the primary objective of effectively managing disorders such as anxiety and depression. These models, in conjunction with mobile health (mHealth) applications and ubiquitous technology, offer innovative approaches to the diagnosis, monitoring, and treatment of mental health disorders. This chapter offers a thorough examination of the most recent research on the efficacy and function of LLMs in the treatment of anxiety and depression.

3.6.1. The Potential of Large Language Models

Language models like GPT-4, developed by OpenAI, have shown impressive abilities in natural language processing. These capabilities can be utilised to enhance mental health care, as suggested by Obradovich et al. [63]. These models have been developed to comprehend and produce text that resembles human language, making them well-suited for a range of applications in the field of psychiatry and psychology. These applications include therapeutic chatbots, diagnostic tools, and personalized treatment plans. LLMs have the potential to greatly improve access to mental health care. The study conducted by Obradovich et al. [63] emphasizes the potential of LLMs in enhancing data collection and developing novel therapeutic tools. This can greatly improve the accessibility of mental health services, particularly in areas that lack sufficient resources. Therapeutic chatbots powered by LLMs offer immediate support and resources to individuals dealing with anxiety or depression. This has the potential to lessen the workload on mental health professionals and enhance patient outcomes.

In their publication, Obradovich et al. [63] delve into the incorporation of LLMs into psychiatric care, emphasizing their significant capacity for transformation. LLMs have the potential to improve diagnostic accuracy, enable personalized care, and simplify administrative processes. They also highlight the capabilities of LLMs, emphasizing their ability to process patient information, streamline therapy sessions, and aid in complex diagnostic problem-solving. This capability has the potential to greatly enhance the accuracy of diagnoses and speed up the delivery of appropriate treatments for patients. Furthermore, LLMs possess the capability to identify patterns and trends in patient data that may not be immediately obvious to human clinicians. The development of more informed and effective treatment strategies can be a potential outcome of this. Nevertheless, the authors highlight the difficulties, including computational requirements and ethical considerations, that require the creation of practical frameworks to guarantee their secure implementation.

A study conducted by Kalisperakis et al. [43] delves into the utilisation of digital phenotypes derived from smartwatches to forecast fluctuations in symptoms among individuals with psychotic disorders. The study primarily examined psychosis, but its methodology and findings have significant implications for the management of anxiety and depression. The study employed smartwatches to gather ongoing biometric data, including heart rate variability and activity levels, over an extended duration. The data was subsequently analysed using LLMs to forecast alterations in the patients’ positive and negative symptoms. The study’s findings suggest a link between elevated heart rate during wakefulness and sleep and heightened positive psychopathology. Conversely, reduced heart rate variability appears to be connected to increased negative psychopathology [43]. This study explores the potential of digital phenotyping tools in tracking physiological signs of anxiety and depression, which could have significant implications for the treatment of these diseases. Regular monitoring and analysis can help identify early signs of worsening symptoms, allowing for timely interventions and personalized adjustments to therapy.

The latest work by Stavrianos et al. [85] delves into the integration of speech emotion recognition (SER) in homecare platforms. The significance of effective computing in improving patient care is highlighted. This research provides a detailed description of the design and implementation of an emotional recognition service, which is a crucial component of an all-inclusive electronic homecare management system. The researchers assess various methods, such as machine learning and deep learning algorithms, for voice analysis to showcase the precision of identifying and interpreting patient emotions. Given the significant impact of emotional states on anxiety and depression, it is crucial to effectively manage these disorders. Integrating Social and Emotional Robotics (SER) into homecare systems allows healthcare personnel to continuously monitor the emotional well-being of patients. This enables a prompt response to emotional discomfort and allows for adjustments in treatment strategies, if needed.

Zlatintsi et al. [97] introduce the E-Prevention system, a sophisticated support tool developed for the purpose of monitoring and preventing relapse in individuals diagnosed with psychotic disorders. This system uses data from wearables and video captures to offer thorough monitoring and analysis. The incorporation of multiple data sources highlights the significance of improving tools for managing mental health. The study demonstrates the potential for enhancing the precision and dependability of mental health interventions by integrating LLMs with other technologies [97]. Gallos et al. [30] investigate how mHealth technologies are used in metropolitan regions to improve public health and well-being. The paper emphasizes how important mHealth tools are for helping people with mental health disorders like anxiety and sadness. These devices provide insightful analysis of patients’ emotional and physical well-being as they gather location and health-related information to track vital signs and physical activity. The integration of mobile health (mHealth) and low-cost, low-maintenance (LLMs) devices exemplify a pragmatic method for improving mental health care using technology [30].

In their study, Maglogiannis et al. [54] present a sophisticated cloud-based platform that has been specifically developed to efficiently monitor individuals with psychotic disorders. This platform utilises cloud computing to store and analyse data from wearable devices and home-based video recordings. Utilising a cloud-based approach allows for efficient processing of large volumes of data, facilitating real-time monitoring and analysis. Continuous surveillance and sophisticated data analysis may provide significant observations about the patient’s state, allowing for timely intervention and customized therapy. The potential advantages of LLMs in the management of anxiety and depression are substantial, however, it is essential to acknowledge and tackle the associated risks and difficulties. LLMs have a notable inclination to generate outcomes that are both unexpected and non-deterministic, which is a significant concern to consider. Obradovich et al. [63] performed research which found that LLMs have the capacity to provide replies that are factually inaccurate or illogical. This might cause difficulties, especially in therapeutic environments where precision is essential. For example, an LLM might possibly provide counsel that is unsuitable or inaccurate, with substantial consequences for those experiencing anxiety or depression. Ensuring the safety and effectiveness of LLM outcomes in mental health treatment requires prioritizing their reliability and correctness.

3.6.2. Examining the Ethical Considerations and Potential Biases

When incorporating LLMs (Linguistic Language Models) in the field of psychiatry or psychology, it is of utmost importance to give priority to ethical issues. Obradovich et al. [63] emphasize the significance of using responsible AI methodologies to tackle possible hazards, such as prejudice, privacy violations, and the reinforcement of stereotypes. LLMs undergo training using large datasets that may have inherent biases, which might manifest in the outcomes they generate. Addressing these prejudices and ensuring that LLMs provide treatment that is fair and unbiased is crucial for maintaining the integrity of mental health services.

Concerns have arisen about the integration of LLMs into therapeutic practice, particularly with potential dependence and the need for human supervision. Obradovich et al. [63] performed a recent study that elucidated the potential hazards of excessive reliance on artificial intelligence by practitioners. This excessive dependence has the capacity to hinder their own understanding and discernment, posing a significant risk. In order to guarantee the safety of patients, it is necessary for human professionals to constantly monitor and intervene in any unforeseen actions performed by LLMs. Establishing protocols for human oversight and intervention is crucial in order to ensure the appropriate use of Language and Learning Models (LLMs) in the field of psychiatry.

3.6.3. Prospects for the Future and Suggestions for Improvement

Further investigation and experimentation are required to fully use the capabilities of LLMs in the treatment of anxiety and depression. Several crucial aspects need attention. The analyzed research highlights the substantial influence of LLMs in alleviating anxiety and depression. The authors provide information that supports the benefits of digital phenotyping, enhanced monitoring systems, mHealth apps, and cloud-based platforms. However, there are still significant challenges that must be addressed, such as ensuring data privacy, addressing computational demands, and establishing ethical principles. It is important to focus on resolving these problems via more research in order to maximize the benefits of LLMs in the realm of mental health treatment.

Ensuring the reliability and accuracy of LLM outcomes is crucial for their safe integration into mental health treatment. This entails enhancing training datasets, implementing rigorous validation processes, and devising techniques to detect and correct faulty outputs. Enhancing the dependability of LLMs may enhance their usefulness as tools for both doctors and patients. Ensuring that LLMs adhere to ethical standards and safeguard patient privacy is of utmost importance. The methodology entails creating structures to ensure the ethical deployment of AI, recognizing and reducing biases in training data, and establishing robust privacy protections. Obradovich et al. propose the use of the Biological-Psychological, Economic, and Social (BPES) model as a comprehensive framework to ensure the ethical and secure implementation of artificial intelligence in the field of mental health therapy [63].

The integration of diverse data sources, including physiological signals captured by wearable devices and emotional cues extracted from speech, has the capacity to enhance the effectiveness of LLMs in addressing anxiety and depression. The research undertaken by Kalisperakis et al. [43], Stavrianos et al. [85], and Zlatintsi et al. [97] emphasizes the need of using various data sources to get a thorough understanding of people’s mental well-being. Additional research should focus on creating integrated systems that use diverse data sources to improve the range and personalization of healthcare. To successfully use LLMs in the area of psychiatry, it is essential to promote multidisciplinary cooperation among AI researchers, clinicians, ethicists, and policymakers. Through collaboration, we can guarantee that the creation and execution of LLMs are directed by the knowledge and skills of medical professionals, ethical factors, and regulatory criteria. By fostering a collaborative mentality, the mental health sector may effectively harness the benefits of artificial intelligence while simultaneously mitigating any possible hazards.

The use of expansive language models in the realm of mental health care has significant promise in enhancing the treatment of anxiety and depression. LLMs may increase access to healthcare, boost diagnosis accuracy, and allow personalized therapy by using their sophisticated natural language processing capabilities. However, the use of these technologies poses substantial hazards, including their unexpected characteristics, ethical concerns, and the need for robust human oversight. To fully use the advantages of LLMs in the area of psychiatry, future research should focus on enhancing the dependability of models, addressing ethical and privacy concerns, integrating multimodal data, and fostering multidisciplinary cooperation. To overcome these challenges, the mental health sector may use AI technology to improve the quality and accessibility of treatment for patients struggling with anxiety and despair.

3.6.4. Enhancing the Accuracy and Efficiency of Psychiatric and Psychological Care

The integration of artificial intelligence (AI) into the field of mental health has opened new avenues for the detection, diagnosis, and treatment of anxiety and depression. Particularly, large language models (LLMs) and multimodal approaches have shown promise in. This comprehensive review explores the current landscape of LLMs in managing anxiety and depression, drawing insights from recent research studies.

Squires et al. conducted a comprehensive assessment on the use of deep learning and machine learning in psychiatry, focusing mostly on depression [83]. The study emphasizes the transformative potential of artificial intelligence (AI) in revolutionizing the diagnosis and treatment of mental health conditions. According to the experts, deep learning algorithms have the ability to analyze large amounts of data and identify complex patterns that may provide a more advanced understanding of mood disorders. Several challenges, including patient privacy issues, the complexity of understanding the models, and their incompatibility with existing healthcare infrastructure, have hindered their extensive use in clinical settings [83].

Xie et al. examined the potential of using multimodal data fusion in their study on anxiety and depression [94]. Their paper proposes a model that combines Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks to integrate various types of input, including text, audio, and physiological signals. This approach provides a comprehensive assessment of the patient’s mental well-being by integrating the advantages of many modalities, hence enhancing the precision of diagnosis. Xie et al. found that multimodal fusion is much more successful than standard single-modality techniques in clinical diagnosis [94].

To get a comprehensive understanding of emotional analytics using multimodal big data, refer to the work of Shoumy et al. [78]. Affective computing systems may enhance their ability to recognize and react to emotional states by incorporating text, auditory, visual, and physiological data. To enhance the development of AI systems for mental health, it is crucial to use a multimodal approach. This technique enables a comprehensive and precise understanding of patients’ emotions and behaviours. Shoumy et al. suggest that these systems have the capacity to fundamentally revolutionize emotional computing and its applications in mental health therapy [78]. Xiao et al. and Zhang et al. performed thorough evaluations of the progress and use of large language models (LLMs) in the domains of mental health and other medical disciplines [93,95]. Xiao et al. examine the ability of LLMs to evaluate and understand medical literature, with the goal of improving clinical decision-making and patient care [93]. Xiao et al. suggest that the use of LLMs capable of analyzing multimodal data might enhance patient assessments, making them more thorough and extensive [93].

Zhang et al. explore recent advancements in multimodal large language models (MM-LLMs) that integrate both visual and textual information to enhance the performance of the models [95]. The authors highlight the potential use of MM-LLMs in healthcare environments, namely for the purposes of diagnosis and treatment planning. The integration of multiple data sources with MM-LLMs may provide crucial and contextually relevant insights for tackling complex disorders like anxiety and depression [95]. Chen et al. examine the efficacy of several tuning procedures for LLMs in the field of medical multimodal analysis [19]. The specific challenges posed by medical data include high dimensionality and heterogeneity. Their research focuses on optimizing LLMs to effectively address these concerns. The result validates prior discoveries that tailored tuning strategies significantly improve LLM performance, making them more appropriate for use in clinical settings [19]. In the realm of mental health, where timely and accurate identification is crucial, this has particular significance [19]. Wu et al. propose LoRA-SP, a simple method for adjusting partial parameters to optimize LLMs [92]. The objective of this method is to enhance resource efficiency while maintaining model performance. According to Wu et al., the authors demonstrate that LoRA-SP has the potential to enhance the practicality of LLMs in healthcare by decreasing processing expenses and improving scalability [92].

Wang et al. thoroughly examine the reasoning abilities of multimodal large language models in their comprehensive study [91]. To enhance the accuracy and depth of their analysis, researchers explore the incorporation and utilization of various types of data, such as text, images, and audio, inside these models. This study focuses on recent advancements in multimodal cognition and their potential applications in addressing mental health conditions, such as depression and anxiety. The authors argue that enhancing the reasoning abilities of LLMs is crucial for developing more efficient and reliable AI systems in the field of healthcare [91].

Shajari et al. analyze the emergence of wearable sensors fueled by artificial intelligence in the field of digital health technologies in their 2023 evaluation [77]. These sensors can monitor physiological data in real-time to assess mental health disorders. The authors assert that the combination of artificial intelligence and wearable technology has potential for advancing early diagnosis and personalized treatment of depression and anxiety. The constant collection and analysis of data facilitated by these sensors may enable more preventive and proactive mental health therapy [77]. Spitzer et al. and Vallée discuss the concept of digital twins in the field of precision mental health [84,90]. Digital twins are virtual replicas of patients that use real-time data to simulate and predict health consequences. Spitzer et al. emphasize the capacity of digital twins to enhance precision mental health by customizing therapy plans and monitoring individuals’ advancements [84]. Digital twins have the capacity to revolutionize mental health care by enabling more accurate and customized treatments, as elaborated by Vallée, who provides a deeper exploration of the role and actuality of digital twins in personalized medicine [90].

The 2019 Global Burden of Disease Study (GBD 2019 Mental disorders Collaborators) incorporates a comprehensive assessment of the worldwide effects of mental illnesses [31]. The study demonstrates that mental health conditions, such as anxiety and depression, have a significant impact on global health. The 2019 report by the Mental disorders Collaborators emphasizes the need to find efficient and adaptable solutions, such as AI and LLMs, to address the growing burden of mental disorders [31].

Figure 5. Mindmap diagram illustrating the application of LLMs in the treatment of anxiety and depression.

Figure 5. Mindmap diagram illustrating the application of LLMs in the treatment of anxiety and depression.

Figure 6. Trends in Research Focus Areas Over Time.

Figure 6. Trends in Research Focus Areas Over Time.

4. Evaluation of AI Tools—Case Studies

5. Discussion, Challenges and Limitations

6. Conclusions and Future Orientations

References

1. Abd-alrazaq A, Alhuwail D, Schneider J, Toro CT, Ahmed A, Alzubaidi M, Househ M. The performance of artificial intelligence-driven technologies in diagnosing mental disorders: an umbrella review. Npj Digital Medicine. 2022;5(1):1–12. [CrossRef]
2. Abd-alrazaq A, AlSaad R, Aziz S, Ahmed A, Denecke K, Househ M, Sheikh J. Wearable Artificial Intelligence for Anxiety and Depression: Scoping Review. J Med Internet Res. 2023;25(1). [CrossRef]
3. Ahmed A, Aziz S, Toro CT, Alzubaidi M, Irshaidat S, Serhan HA, Househ M. Machine learning models to detect anxiety and depression through social media: A scoping review. Comput Methods Programs Biomed Update. 2022;2:100066. [CrossRef]
4. Ahmed A, Hassan A, Aziz S, Abd-alrazaq AA, Ali N, Alzubaidi M, Househ M. Chatbot features for anxiety and depression: A scoping review. Health Inform J. 2023;29(1):146045822211467. [CrossRef]
5. Ahorsu DK, Lin CY, Imani V, Saffari M, Griffiths MD, Pakpour AH. The Fear of COVID-19 Scale: Development and Initial Validation. Int J Ment Health Addict. 2020. [CrossRef]
6. Alberts NM, Leisenring WM, Flynn JS, Whitton J, Gibson TM, Jibb L, Armstrong GT. Wearable Respiratory Monitoring and Feedback for Chronic Pain in Adult Survivors of Childhood Cancer: A Feasibility Randomized Controlled Trial From the Childhood Cancer Survivor Study. JCO Clin Cancer Inform. 2020;4:1014–1026. [CrossRef]
7. Andrews G, Cuijpers P, Craske MG, McEvoy P, Titov N. Computer Therapy for the Anxiety and Depressive Disorders Is Effective, Acceptable and Practical Health Care: A Meta-Analysis. PLoS ONE. 2010;5(10). [CrossRef]
8. Austin SF, Frøsig A, Buus N, Lincoln T, von Malachowski A, Schlier B, Simonsen E. Service User Experiences of Integrating a Mobile Solution (IMPACHS) Into Clinical Treatment for Psychosis. Qual Health Res. 2021;31(5):942–954. [CrossRef]
9. Baghaei N, Chitale V, Hlasnik A, Stemmet L, Liang HN, Porter R. Virtual Reality for Supporting the Treatment of Depression and Anxiety: Scoping Review. JMIR Ment Health. 2021;8(9). [CrossRef]
10. Bakker D, Kazantzis N, Rickwood D, Rickard N. A randomized controlled trial of three smartphone apps for enhancing public mental health. Behav Res Ther. 2018;109:75–83. [CrossRef]
11. Balcombe L, de Leo D. Psychological screening and tracking of athletes and digital mental health solutions in a hybrid model of care: Mini review. JMIR Form Res. 2020;4(12). [CrossRef]
12. Balcombe L, De Leo D. An Integrated Blueprint for Digital Mental Health Services Amidst COVID-19. JMIR Ment Health. 2020;7(7). [CrossRef]
13. Balcombe L, De Leo D. Human-Computer Interaction in Digital Mental Health. Informatics. 2022;9(1). [CrossRef]
14. Barua PD, Vicnesh J, Lih OS, Palmer EE, Yamakawa T, Kobayashi M, Acharya UR. Artificial intelligence assisted tools for the detection of anxiety and depression leading to suicidal ideation in adolescents: a review. Cogn Neurodyn. 2024;18(1):1–22. [CrossRef]
15. Bautista J, Liu M, Alvarez M, Schueller SM. Multi-Media Field Test: Cognitive-Behavioral Therapy at Our Fingertips: Sanvello Provides On-Demand Support for Mental Health. Cogn Behav Pract. 2024. [CrossRef]
16. Becker TD, Torous JB. Recent Developments in Digital Mental Health Interventions for College and University Students. Curr Treat Options Psychiatry. 2019;6(3):210–220. [CrossRef]
17. Berenguer C, Baixauli I, Gómez S, Andrés MdEP, De Stasio S. Exploring the Impact of Augmented Reality in Children and Adolescents with Autism Spectrum Disorder: A Systematic Review. Int J Environ Res Public Health. 2020;17(17):1–15. [CrossRef]
18. Chandra S, Mohammadnezhad M, Ward P. Trust and Communication in a Doctor-Patient Relationship: A Literature Review. J Health Commun. 2018;3(3):1–6. [CrossRef]
19. Chen J, Jiang Y, Yang D, Li M, Wei J, Qian Z, Zhang L. Can LLMs’ tuning methods work in the medical multimodal domain? arXiv preprint arXiv:2403.06407.
20. Cheng Y, Wang K, Xu H, Li T, Jin Q, Cui D. Recent developments in sensors for wearable device applications. Anal Bioanal Chem. 2021;413(24):6037–6057. [CrossRef]
21. Chou YH, Lin C, Lee SH, Lee YF, Cheng LC. User-Friendly Chatbot to Mitigate the Psychological Stress of Older Adults During the COVID-19 Pandemic: Development and Usability Study. JMIR Form Res. 2024;8.
22. . [CrossRef]
23. Corrigan PW, Mittal D, Reaves CM, Haynes TF, Han X, Morris S, Sullivan G. Mental health stigma and primary health care decisions. Psychiatry Res. 2014. [CrossRef]
24. De Witte NAJ, Joris S, Van Assche E, Van Daele T. Technological and Digital Interventions for Mental Health and Wellbeing: An Overview of Systematic Reviews. Front Digit Health. 2021;3. [CrossRef]
25. Ding NB. Anxiety disorders and phobias: A cognitive perspective. Health Soc Work. 1988;13(1):76–77. [CrossRef]
26. Ewbank MP, Cummins R, Tablan V, Bateup S, Catarino A, Martin AJ, Blackwell AD. Quantifying the Association Between Psychotherapy Content and Clinical Outcomes Using Deep Learning. JAMA Psychiatry. 2020;77(1):35. [CrossRef]
27. Firth J, Torous J, Nicholas J, Carney R, Pratap A, Rosenbaum S, Sarris J. The efficacy of smartphone-based mental health interventions for depressive symptoms: a meta-analysis of randomized controlled trials. World Psychiatry. 2017;16(3):287–298. [CrossRef]
28. Fitzpatrick KK, Darcy A, Vierhile M. Delivering Cognitive Behavior Therapy to Young Adults With Symptoms of Depression and Anxiety Using a Fully Automated Conversational Agent (Woebot): A Randomized Controlled Trial. JMIR Ment Health. 2017;4(2). [CrossRef]
29. Fulmer R, Joerin A, Gentile B, Lakerink L, Rauws M. Using Psychological Artificial Intelligence (Tess) to Relieve Symptoms of Depression and Anxiety: Randomized Controlled Trial. JMIR Ment Health. 2018;5(4). [CrossRef]
30. Gallos P, Menychtas A, Panagopoulos C, Kaselimi M, Temenos A, Rallis I, Doulamis A, Doulamis N, Bimpas M, Aggeli A, Protopapadakis E, Sardis E, Maglogiannis I. Using mHealth technologies to promote public health and well-being in urban areas with blue-green solutions. Stud Health Technol Inform. 2022;295:566-569. [CrossRef]
31. GBD 2019 Mental Disorders Collaborators. Global, regional, and national burden of 12 mental disorders in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet Psychiatry. 2022;9(2):137-150.
32. Gelfgren S, Ineland J, Cocq C. Social media and disability advocacy organizations caught between hopes and realities. Disabil Soc. 2022;37(7):1085–1106. [CrossRef]
33. Glenn T, Monteith S. New Measures of Mental State and Behavior Based on Data Collected From Sensors, Smartphones, and the Internet. Curr Psychiatry Rep. 2014;16(12):523. [CrossRef]
34. Goldberg SB, Flemotomos N, Martinez VR, Tanana MJ, Kuo PB, Pace BT, Atkins DC. Machine learning and natural language processing in psychotherapy research: Alliance as example use case. J Couns Psychol. 2020;67(4):438–448. [CrossRef]
35. Gomes N, Pato M, Lourenço AR, Datia N. A Survey on Wearable Sensors for Mental Health Monitoring. Sensors. 2023;23(3):1–16. [CrossRef]
36. González Ramírez ML, García Vázquez JP, Rodríguez MD, Padilla-López LA, Galindo-Aldana GM, Cuevas-González D. Wearables for Stress Management: A Scoping Review. Healthcare. 2023;11(17):2369. [CrossRef]
37. Grimm B, Talbot B, Larsen L. PHQ-V/GAD-V: Assessments to identify signals of depression and anxiety from patient video responses. Appl Sci. 2022;12(18):9150. [CrossRef]
38. Healthline. Anxiety Chest Pain: Symptoms, Home Remedies, and More. Retrieved December 3, 2019, from https://www.healthline.com/health/anxiety/anxiety-chest-pain#see-a-doctor.
39. Inkster B, Sarda S, Subramanian V. An Empathy-Driven, Conversational Artificial Intelligence Agent (Wysa) for Digital Mental Well-Being: Real-World Data Evaluation Mixed-Methods Study. JMIR MHealth UHealth. 2018;6(11). [CrossRef]
40. Ito A, Hiyoshi F, Kanie A, Maruyama A, Oba MS, Kito S. Feasibility Study of Virtual Reality–Based Cognitive Behavioral Therapy for Patients With Depression: Protocol for an Open Trial and Therapeutic Intervention. JMIR Res Protoc. 2023;12. [CrossRef]
41. Ivanov VM, Krivtsov AM, Strelkov SV, Smirnov AY, Shipov RY, Grebenkov VG, Koskin VS. Practical Application of Augmented/Mixed Reality Technologies in Surgery of Abdominal Cancer Patients. J Imaging. 2022;8(7). [CrossRef]
42. Jingili N, Oyelere SS, Nyström MBT, Anyshchenko L. A systematic review on the efficacy of virtual reality and gamification interventions for managing anxiety and depression. Front Digit Health. 2023;5. [CrossRef]
43. Kalisperakis E, Karantinos T, Lazaridi M, Garyfalli V, Filntisis PP, Zlatintsi A, Efthymiou N, Mantas A, Mantonakis L, Mougiakos T, Maglogiannis I, Tsanakas P, Maragos P, Smyrnis N. Smartwatch digital phenotypes predict positive and negative symptom variation in a longitudinal monitoring study of patients with psychotic disorders. Front Psychiatry. 2023;14:1024965. [CrossRef]
44. Karkosz S, Szymański R, Sanna K, Michałowski J. Effectiveness of a Web-based and Mobile Therapy Chatbot on Anxiety and Depressive Symptoms in Subclinical Young Adults: A Randomized Controlled Trial (Preprint). JMIR Form Res. 2023;8. [CrossRef]
45. Kawadler JM, Hemmings NR, Ponzo S, Morelli D, Bird G, Plans D. Effectiveness of a smartphone app (Biobase) for reducing anxiety and increasing mental well-being: Pilot feasibility and acceptability study. JMIR Form Res. 2020;4(11). [CrossRef]
46. Kleine AK, Lermer E, Cecil J, Heinrich A, Gaube S. Advancing mental health care with AI-enabled precision psychiatry tools: A patent review. Comput Human Behav Rep. 2023;12(March):100322. [CrossRef]
47. Ku WL, Min H. Evaluating Machine Learning Stability in Predicting Depression and Anxiety Amidst Subjective Response Errors. Healthcare. 2024;12(6):625. [CrossRef]
48. Lattie EG, Nicholas J, Knapp AA, Skerl JJ, Kaiser SM, Mohr DC. Opportunities for and Tensions Surrounding the Use of Technology-Enabled Mental Health Services in Community Mental Health Care. Adm Policy Ment Health Ment Health Serv Res. 2020;47(1):138–149. [CrossRef]
49. Lederbogen F, Kirsch P, Haddad L, Streit F, Tost H, Schuch P, Meyer-Lindenberg A. City living and urban upbringing affect neural social stress processing in humans. Nature. 2011;474(7352):498–501. [CrossRef]
50. Lee EE, Torous J, De Choudhury M, Depp CA, Graham SA, Kim HC, Jeste DV. Artificial Intelligence for Mental Health Care: Clinical Applications, Barriers, Facilitators, and Artificial Wisdom. Biol Psychiatry Cogn Neurosci Neuroimaging. 2021;6(9):856–864. [CrossRef]
51. Lovejoy CA, Buch V, Maruthappu M. Technology and mental health: The role of artificial intelligence. Eur Psychiatry. 2019;55(January):1–3. [CrossRef]
52. Lu L, Zhang J, Xie Y, Gao F, Xu S, Wu X, Ye Z. Wearable health devices in health care: Narrative systematic review. JMIR MHealth UHealth. 2020;8(11). [CrossRef]
53. Lundin RM, Yeap Y, Menkes DB. Adverse Effects of Virtual and Augmented Reality Interventions in Psychiatry: Systematic Review. JMIR Ment Health. 2023;10. [CrossRef]
54. Maglogiannis I, Zlatintsi A, Menychtas A, Papadimatos D, Filntisis PP, Efthymiou N, Retsinas G, Tsanakas P, Maragos P. An intelligent cloud-based platform for effective monitoring of patients with psychotic disorders. Proceedings of the International Conference on Artificial Intelligence Applications and Innovation (AIAI-2020). [CrossRef]
55. Mehta A, Niles AN, Vargas JH, Marafon T, Couto DD, Gross JJ. Acceptability and Effectiveness of Artificial Intelligence Therapy for Anxiety and Depression (Youper): Longitudinal Observational Study. J Med Internet Res. 2021;23(6). [CrossRef]
56. Mercer SW, Reynolds WJ. Empathy and quality of care. Br J Gen Pract. 2002;52(SUPPL.):9–12.
57. Mohr DC, Burns MN, Schueller SM, Clarke G, Klinkman M. Behavioral Intervention Technologies: Evidence review and recommendations for future research in mental health. Gen Hosp Psychiatry. 2013;35(4):332–338. [CrossRef]
58. Morales A, Barbosa M, Morás L, Cazella SC, Sgobbi LF, Sene I, Marques G. Occupational Stress Monitoring Using Biomarkers and Smartwatches: A Systematic Review. Sensors. 2022;22(17):6633. [CrossRef]
59. Muñoz S, Iglesias C. Detection of the severity level of depression signs in text combining a feature-based framework with distributional representations. Appl Sci. 2023;13(21):11695. [CrossRef]
60. Top of Form.
61. Bottom of Form.
62. Nemesure MD, Heinz MV, Huang R, Jacobson NC. Predictive modeling of depression and anxiety using electronic health records and a novel machine learning approach with artificial intelligence. Sci Rep. 2021;11(1):1–9. [CrossRef]
63. Newson JJ, Thiagarajan TC. Assessment of Population Well-Being With the Mental Health Quotient (MHQ): Development and Usability Study. JMIR Ment Health. 2020;7(7). [CrossRef]
64. Nicosia J, Wang B, Aschenbrenner AJ, Sliwinski MJ, Yabiku ST, Roque NA, Hassenstab J. To BYOD or not: Are device latencies important for bring-your-own-device (BYOD) smartphone cognitive testing? Behav Res Methods. 2023;55(6):2800–2812. [CrossRef]
65. Obradovich N, Khalsa SS, Khan WU, Suh J, Perlis RH, Ajilore O, Paulus MP. Opportunities and risks of large language models in psychiatry. Nat Ment Health. 2024. [CrossRef]
66. Odgers CL, Jensen MR. Annual Research Review: Adolescent mental health in the digital age: facts, fears, and future directions. J Child Psychol Psychiatry. 2020;61(3):336–348. [CrossRef]
67. Oh J, Kim M, Park H, Oh H. Are you depressed? Analyze user utterances to detect depressive emotions using DistilBERT. Appl Sci. 2023;13(10):6223. [CrossRef]
68. Pentina I, Xie T, Hancock T, Bailey A. Consumer–machine relationships in the age of artificial intelligence: Systematic literature review and research directions. Psychol Mark. 2023;40(8):1593–1614. [CrossRef]
69. Ponzo S, Morelli D, Kawadler JM, Hemmings NR, Bird G, Plans D. Efficacy of the Digital Therapeutic Mobile App BioBase to Reduce Stress and Improve Mental Well-Being Among University Students: Randomized Controlled Trial. JMIR MHealth UHealth. 2020;8(4). [CrossRef]
70. Rani K, Vishnoi H, Mishra M. A Mental Health Chatbot Delivering Cognitive Behavior Therapy and Remote Health Monitoring Using NLP And AI. 2023 Int Conf Disrupt Technol ICDT. 2023;313–317. [CrossRef]
71. Rappaport LM, Jerome E, Van Ameringen M, Whittal M, McLean CP. North American open-label 16-week trial of the MindShift smartphone app for adult anxiety. J Mood Anxiety Disord. 2023;4(October):100036. [CrossRef]
72. Romanovskyi O, Pidbutska N, Knysh A. Elomia chatbot: The effectiveness of artificial intelligence in the fight for mental health. CEUR Workshop Proc. 2021;2870:1215–1224.
73. Rowe NC, Chan AL. Rating whole-body suspiciousness factors in automated surveillance of a public area. Proc 2011 Int Conf Image Process Comput Vis Pattern Recognit IPCV 2011. 2011;1:317–322.
74. Sadeh-Sharvit S, Camp T. Del, Horton SE, Hefner JD, Berry JM, Grossman E, Hollon SD. Effects of an Artificial Intelligence Platform for Behavioral Interventions on Depression and Anxiety Symptoms: Randomized Clinical Trial. J Med Internet Res. 2023;25(8). [CrossRef]
75. Sandhu DE. Compute Depression and Anxiety Among Students in Pakistan, Using Machine Learning. Lahore Garrison Univ Res J Comput Sci Inf Technol. 2022;6(04):4–11. [CrossRef]
76. Schröder D, Wrona KJ, Müller F, Heinemann S, Fischer F, Dockweiler C. Impact of virtual reality applications in the treatment of anxiety disorders: A systematic review and meta-analysis of randomized-controlled trials. J Behav Ther Exp Psychiatry. 2023;81:101893. [CrossRef]
77. Schueller SM, Neary M, O’Loughlin K, Adkins EC. Discovery of and Interest in Health Apps Among Those With Mental Health Needs: Survey and Focus Group Study. J Med Internet Res. 2018;20(6). [CrossRef]
78. Schyff EL, Ridout B, Amon KL, Forsyth R, Campbell AJ. Providing Self-Led Mental Health Support Through an Artificial Intelligence–Powered Chat Bot (Leora) to Meet the Demand of Mental Health Care. J Med Internet Res. 2023;25. [CrossRef]
79. Shajari S, Kuruvinashetti K, Komeili A, Sundararaj U. The emergence of AI-based wearable sensors for digital health technology: a review. Sensors (Basel). 2023;23(23):9498.
80. Shoumy NJ, Ang LM, Seng KP, Rahaman DM, Zia T. Multimodal big data affective analytics: A comprehensive survey using text, audio, visual and physiological signals. J Netw Comput Appl. 2020;149:102447.
81. Sharma G, Schlosser L, Jones BDM, Blumberger DM, Gratzer D, Husain MO, Husain MI. Brief App-Based Cognitive Behavioral Therapy for Anxiety Symptoms in Psychiatric Inpatients: Feasibility Randomized Controlled Trial. JMIR Form Res. 2022;6(11). [CrossRef]
82. Sigurðardóttir S, Helgadóttir FD, Menzies RE, Sighvatsson MB, Menzies RG. Improving adherence to a web-based cognitive-behavioural therapy program for social anxiety with group sessions: A randomised control trial. Internet Interv. 2022;28:100535. [CrossRef]
83. Skapinakis P, Bellos S, Oikonomou A, Dimitriadis G, Gkikas P, Perdikari E, Mavreas V. Depression and its relationship with coping strategies and illness perceptions during the Covid-19 lockdown in Greece: A cross-sectional survey of the population. Depress Res Treat. 2020. [CrossRef]
84. Smith EN, Santoro E, Moraveji N, Susi M, Crum AJ. Integrating wearables in stress management interventions: Promising evidence from a randomized trial. Int J Stress Manag. 2020;27(2):172–182. [CrossRef]
85. Squires M, Tao X, Elangovan S, Gururajan R, Zhou X, Acharya UR, Li Y. Deep learning and machine learning in psychiatry: a survey of current progress in depression detection, diagnosis and treatment. Brain Inform. 2023;10(1):10.
86. Spitzer M, Dattner I, Zilcha-Mano S. Digital twins and the future of precision mental health. Front Psychiatry. 2023;14:1082598.
87. Stavrianos P, Pavlopoulos A, Maglogiannis I. Enabling speech emotional intelligence as a service in homecare platforms. In: EAI/Springer Innovations in Communication and Computing. Springer; 2022:119-144. [CrossRef]
88. Su L. Access and Engagement of MoodTools, an mHealth Application for Depression (Georgia: Georgia State University). 2020. Retrieved from http://ezproxy.haifa.ac.il/login?url=https://search.ebscohost.com/login.aspx?direct=true&db=ddu&AN=3817E7D915865FCD&site=eds-live&scope=site.
89. Suso-Ribera C, Fernández-Álvarez J, García-Palacios A, Hoffman HG, Bretón-López J, Baños RM, Botella C. Virtual Reality, Augmented Reality, and in Vivo Exposure Therapy: A Preliminary Comparison of Treatment Efficacy in Small Animal Phobia. Cyberpsychol Behav Soc Netw. 2019;22(1):31–38. [CrossRef]
90. Tausczik YR, Pennebaker JW. The psychological meaning of words: LIWC and computerized text analysis methods. J Lang Soc Psychol. 2010;29(1):24–54. [CrossRef]
91. Oshevwe Y, Oluwatoyin AF, Paschal CM, Chinaemelum OC, Tolulope SO. The role of technology in enhancing mental health advocacy: a systematic review. Int J Appl Res Soc Sci. 2024;6(1):37–50. [CrossRef]
92. Vallée A. Envisioning the future of personalized medicine: role and realities of digital twins. J Med Internet Res. 2024;26.
93. Wang Y, Chen W, Han X, Lin X, Zhao H, Liu Y, Yang H. Exploring the reasoning abilities of multimodal large language models (mllms): A comprehensive survey on emerging trends in multimodal reasoning. arXiv preprint arXiv:2401.06805.
94. Wu Y, Xiang Y, Huo S, Gong Y, Liang P. LoRA-SP: streamlined partial parameter adaptation for resource efficient fine-tuning of large language models. In: Third International Conference on Algorithms, Microchips, and Network Applications (AMNA 2024). SPIE; 2024:488-496.
95. Xiao H, Zhou F, Liu X, Liu T, Li Z, Liu X, Huang X. A comprehensive survey of large language models and multimodal large language models in medicine. arXiv preprint arXiv:2405.08603.
96. Xie W, Wang C, Lin Z, Luo X, Chen W, Xu M, Cheng M. Multimodal fusion diagnosis of depression and anxiety based on CNN-LSTM model. Comput Med Imaging Graph. 2022;102:102128.
97. Zadgaonkar A, Keskar R, Kakde O. Towards a machine learning model for detection of dementia using lifestyle parameters. Appl Sci. 2023;13(19):10630. [CrossRef]
98. Zhang W, Lin T, Liu J, Shu F, Li H, Zhang L, Zhuang Y. HyperLLaVA: Dynamic visual and language expert tuning for multimodal large language models. arXiv preprint arXiv:2403.13447.
99. Zlatintsi A, Filntisis PP, Garou C, Efthymiou N, Maragos P, Menychtas A, Maglogiannis I, Tsanakas P, Sounapoglou T, Kalisperakis E. E-Prevention: Advanced support system for monitoring and relapse prevention in patients with psychotic disorders analyzing long-term multimodal data from wearables and video captures. Sensors (Basel). 2022;22(19):7544. [CrossRef]
100. Anxietycanada.com. New MindShift™ CBT App Gives Canadians Free Anxiety Relief. 2024 May 10. Available from: https://www.anxietycanada.com/articles/new-mindshift-cbt-app-gives-canadians-free-anxiety-relief/.
101. Brooks M. AI Tool Flags Anxiety, Depression in Healthcare Workers. Medscape. 2023. Available from: https://www.medscape.com/viewarticle/998047?form=fpf.
102. Calm.com. Calm your mind. Change your life. 2024. Available from: https://www.calm.com/.
103. Callahan L. Lifeline Ally. Theresanaiforthat.com. 2023. Available from: https://theresanaiforthat.com/gpt/lifeline-ally/.
104. Chrome Web Store. Breathhh: Workplace Wellbeing Companion. 2024. Available from: https://chromewebstore.google.com/detail/breathhh-workplace-wellbe/niipedbmjiopjpmjcpigiflabghcckeo/reviews.
105. Crunchbase. Ladder. 2024. Available from: https://www.crunchbase.com/organization/ladder-433e.
106. d3.harvard.edu. Ginger.io: An app that monitors your mental health. 2018. Available from: https://d3.harvard.edu/platform-digit/submission/ginger-io-an-app-that-monitors-your-mental-health/.
107. https://thataicollection.com/en/application/deepwander/.
108. Headspace for Organizations. Mental healthcare for every moment [Internet]. Available from: https://organizations.headspace.com/.
109. Kintsugihealth.com. “I’m fine. 2024. Available from: https://www.kintsugihealth.com/.
110. Liu L. What makes Inflection’s Pi a great companion chatbot. Medium. 2023. Available from: https://medium.com/@lindseyliu/what-makes-inflections-pi-a-great-companion-chatbot-8a8bd93dbc43.
111. Meruhealth.com. Better mental health starts here. 2023. Available from: https://www.meruhealth.com/.
112. Minddoc.com. Mental Health Starts with You. 2024. Available from: https://minddoc.com/us/en.
113. Mindwell.ai. Your virtual self-care partner. 2024. Available from: https://mindwell.ai/en/treatments/anxiety-treatment.
114. Moodmission.com. Change the way you feel. 2024. Available from: https://moodmission.com/.
115. Moodtools.org. MoodTools. 2024. Available from: https://moodtools.org/.
116. Pacifica-stress-anxiety.en.softonic.com. Sanvello: Anxiety Depression APK for Android. 2024 May 12. Available from: https://pacifica-stress-anxiety.en.softonic.com/android.
117. Psychiatry.org—MoodPath.
118. Rawat M. Best AI Apps For Mental Health (2023). MarkTechPost. 2023. Available from: https://www.marktechpost.com/2023/04/11/best-ai-apps-for-mental-health-2023/.
119. Reachout.com. MoodKit. 2024. Available from: https://au.reachout.com/tools-and-apps/moodkit.
120. Rootd.io. Panic attack and anxiety relief. 2024. Available from: https://www.rootd.io/.
121. Sawh M. Stress wearables: best devices that monitor stress and how they work. Wareable. 2022. Available from: https://www.wareable.com/health-and-wellbeing/stress-monitoring-wearables-explained-7969.
122. Solis E. How Mental Health Apps Are Handling Personal Information. New America. 2024. Available from: https://www.newamerica.org/oti/blog/how-mental-health-apps-are-handling-personal-information/.
123. Space of Mind: Affordable PTSD facilitated peer support. Find a therapist to help you in anonymous, group facilitated peer support.
124. Thehub.io. Moodmetric. 2024. Available from: https://thehub.io/startups/moodmetric.
125. Thomasmore.be. App: Phobos AR. 2024. Available from: https://thomasmore.be/en/exp-zw/mw/app-phobos-ar.
126. University of Tennessee. Artificial Intelligence (AI) Tools. 2024. Available from: https://research.utk.edu/research-integrity/artificial-intelligence-ai-tools/.
127. Fisher Wallace. A Superdevice for Depression and Anxiety. 2024. Available from: https://www.fisherwallace.com/.
128. Happify.com. Overcome negative thoughts, stress & life’s challenges! 2024. Available from: https://www.happify.com/.
129. Samsung. Measure your stress level with Samsung Health. 2024. Available from: https://www.samsung.com/us/support/answer/ANS00080574/.
130. Spire Health. Welcome to Health Tag. 2024. Available from: https://www.spirehealth.com/patient-setup.