Artificial Intelligence and machine learning are emerging technologies which is most possibly leading to next-gen technological advancement. When amalgamating A.I. and counseling in the year 1964 to 1966, the interaction between human beings and the computer chat gave rise to ELIZA. The chatbot was intelligent if you limit your conversation to talking about yourself and your life. Thus, pattern matching and string processing are used for the interaction between humans and computers during that time (Sakthivel, Srihari, et. al., 2018). Technology has developed so much today that face detection using image processing is having multiple scopes today. Enhanced methods like Neural Networks and Deep Neural Network (DNN) are used today to handle complex data and problems (Nixon, et. al., 2022). Using Face detection through the computer vision method we can identify various emotions a person is going through by facial cues appearing in the image.
Face emotion analysis is used to estimate the stress or depression level of anyone (Nicolaou, Zafeiriou, 2019) (Liu, Nguen, 2017). In a study by Babu et al. (2022) emotional analysis reviews social medial data for fears as well as identifying depression using artificial intelligence techniques. The study used social media data which consists of texts, emoticons, and emojis to analyze emotion using various artificial intelligence techniques. As a result, Deep Learning Algorithm using multi-class classification shows high accuracy during emotional analysis. Another study by Cousins, et al. (2022), created a huge database consisting of subjects to perform five mood elicitation tasks. After each activity, audio, as well as video tutorials, were collected which consists of 3D knowledge of facial expression with Kinect (motion sensing input device produced by Microsoft). Cousins, et al (2022) proposed two types of deep-seated (DBN) network models which consist of extracted 2D visual features, and a model rendered in 3D which consists of dynamic features. Thus, experimental results indicate that the method is helpful to identify patent at risk of possible depression (Nixon, et. al., 2022).
Zhao et al. (2018) suggested an artificial intelligence system to monitor depression where new technology is used that can infer a person’s emotions from RF signals. EQ-Radio transmits an RF signal and analyzes its reflections off a person’s body so that the emotional state (happy, sad, etc.) is identified (Zhao et al., 2018). The paper further talks about the use of a heartbeat extraction algorithm for determining the beat-to-beat intervals and exploited this interval emotion recognition which may have paved the way for future explorations. Another study by Mohan et al. (2021) suggested a technique to predict depression state by the combination of the result generated from facial expression and sentiment analysis. Beck Depression Inventory-II (BDI-II) can also be predicted using voice and visual expression (Mohan et al., 2021). In consequence, Meshram, et al. (2020) suggests an Artificial Intelligence based “AI Therapy” for depression counseling.
The inspiration for this write-up came from a visual novel I had come across called Eliza by Zachtronics. It is a visual novel about AI counseling program, the people who develop it, and the people who use it. Anyhow, the use of a fully developed AI which can analyze the client’s texts, facial expression, voice tonality, vocabulary, heart rate, temperature, and even past behaviors as well as records to assess the patient’s emotional distress while also providing medication and therapy is fascinating. Some might argue that we need a human connection to open up in a counseling session but to solve that issues the Mental health counseling company named Eliza (in the visual novel) just kept a person in front of the patient to give them the ‘human connection’ and the person can just recite the script handed by the AI program for it to further analyze the client’s symptoms. The initial patients/ clients in the visual novel did come to a threshold where they showed backlash and demanded the human connection but the heroine in the novel who was kept as a proxy to provide that ‘human touch’ was strictly told to follow the script given by the AI no matter what. Eliza (AI program in the visual novel) was made so in touch with reality that there is no denying it is a possibility in the future. It had its phases in the counseling session which included the Introductory phase (building rapport and psychological state model of the client), Discovery phase (Exploring the client’s problem and beginning to introduce probing questions), Challenge phase (in this phase, the program detects the client’s rationalization and or explanatory strategies), Intervention phase (The program selects one or multiple interventions from its toolkit of cognitive exercises, therapeutic application, and recommendations for prescription medicine) and the last phase is Conclusion (Client is asked to rate the experience and the proxy provider receives the client’s feedback as well as performance-based incentives as determined by Eliza).
Screenshot from the visual novel ‘Eliza’ developed by Zachtronics
Comentarios