Graduates of the Information Technology Programme at the Hong Kong Institute of Vocational Education (IVE) have created a system that uses artificial intelligence (AI) technology to evaluate emotional distress and depression levels, as well as associated risks, by analysing various factors such as the caller’s voice, emotions, and keywords.
Currently, various government departments and social welfare organisations offer mental health helplines and online support services. The “AI Early Depression Screening System” has been specifically developed for individuals seeking emotional assistance and the dedicated counselling volunteers who assist them.
For instance, within this system, counselling volunteers from the organisation can engage in online sessions with those seeking help through the platform. They can access the medical records and past conversation logs of the help-seekers.
As the online conversation begins, the system automatically analyses the caller’s speech tone, voice, and content to gain insights into their emotions. Additionally, the system provides a real-time transcription of the caller’s voice, enabling the identification of specific words used during the conversation.
To train the AI system and analyse the conversation content, the team used the depression research lexicon created by Stanford University, which consists of over 2,000 depression-related keywords. For instance, instances have been observed where single mothers repeatedly use the keyword “useless” to describe themselves. In such cases, the system records the frequency of usage and analyses its association with other words, allowing an estimation to be made that the help seeker may be experiencing emotional distress.
Once the conversation is concluded, the system performs an initial assessment of the social worker’s risk of depression or emotional distress. It displays keywords associated with depression and potential causes of the condition. Additionally, it automatically identifies words related to substance abuse, alcohol, gambling, and self-harm as part of assessing the potential suicidal tendencies of the help-seekers. In conclusion, the system offers mental health guidance and suicide prevention advice while generating reports for further analysis and follow-up by professionals.
A senior lecturer at the Department of Information Technology at IVE (Tsing Yi) and the project’s guiding authority, highlights the convenience and user-friendly nature of the integrated online consultation and intelligent analysis system. The system effectively presents counselling conversation data, aiding in preliminary screening and serving as a reference tool for professional counselling or medical staff. However, he emphasised that the diagnosis and treatment of mental health conditions still require the professional judgment of healthcare and counselling teams.
It is suggested that this system is suitable for mental health institutions, hospitals, clinics, and even insurance companies. The continuation of conversations regularly is encouraged to enable the system to gather more data, thereby improving the accuracy of its analysis. Currently, the system exclusively supports English, but the next objective is to optimise it to accommodate additional languages such as Cantonese and Mandarin.
The system has garnered significant recognition and achievements through participation in various competitions in China and Hong Kong.
IVE’s IT Discipline has developed a comprehensive range of programmes to equip learners with the necessary knowledge and skills to thrive in the fast-paced and ever-evolving career landscape. These programmes cover various domains within the field of Information and Communications Technology (ICT) as well as Multimedia and Entertainment Technology (MET). By offering a wide array of courses, the IT Discipline aims to ensure that learners stay up-to-date with the latest advancements and acquire the relevant expertise required in these dynamic industries.