Digital Health

As the world continues to make strides in artificial intelligence (AI), the need for transparency in the field intensifies. Clear and understandable explanations for the predictions of AI models not only enhances user confidence but also enables effective decision-making. Such explanations are especially crucial in sectors like healthcare where predictions can have significant and sometimes life-changing consequences. A prime example is the diagnosis of cardiovascular diseases based on heart murmurs, where an incorrect or misunderstood diagnosis can have severe implications. The technology, DiagramNet, is designed to offer human-like intuitive explanations for diagnosing cardiovascular diseases from heart sounds. It leverages the human  reasoning processes of abduction and deduction to generate hypotheses of what diseases could have caused the specific heart sound, and to evaluate the hypotheses based on rules.

Timely and adequate rehabilitation is critical in facilitating post-stroke recovery. However, the organisation and delivery of rehabilitation are resource-demanding, and are only available in ~25% of stroke survivors in low-middle-income countries. Therefore, innovative solutions are urgently required to improve access to stroke rehabilitation services. Our team developed a platform to assist remote dissemination of physio- and occupational therapy services. Our platform enables post-stroke rehabilitation services to be delivered using consumer-grade devices (e.g., smartphones and tablet devices) and artificial intelligence technology, thereby breaking many physical and resource barriers to rehabilitation service access. Our platform is designed for hospitals, rehabilitation centres and post-stroke service providers. The system consists of two interfaces – the therapist portal and patient application. Portal system enables remote prescription of exercises and monitor exercise performance and progress, whereas the patient application acts as a “smart therapist” in providing real-time instructions and feedback. The system addresses limitations faced by the conventional phone interview- and video conference style of telemedicine, which mainly relies on self-reporting and subjective recall. Using devices’ built-in camera and relevant Internet of Things (IoTs), we automated the evaluation process using algorithms with key features identified by professional therapists for providing “supervision” through real-time feedback.

Capsule endoscopy is a non-invasive diagnostic method that utilizes a tiny, wireless camera capsule to capture images of the digestive system, which aids in identifying various conditions. Studies indicate that colon cancer is the leading cause of cancer-related deaths among men aged 20-49, and capsule endoscopy is the least invasive examination capable of detecting the early stages of this cancer, alongside other pathologies such as inflammatory bowel disease, ulcers, tumors, and gastro-intestinal bleeding. Without this procedure, these conditions often go undetected and become untreatable by the time noticeable symptoms appear. This technology offer is a remote capsule endoscopy examination, aiming to make colon diagnosis more widespread among people and extend its use to animals, primarily pets like dogs. Regular capsule endoscopy colon screenings are the only way to prevent the growth of undetectable pathologies. Remote technology enables patients to move around or work without needing to stay in a medical facility, allowing diagnoses to be conducted in areas previously unreachable by capsule endoscopy, such as on battlefields, in underdeveloped regions, or even by astronauts in space. The system includes an artificial intelligence module that automatically identifies and suggests potential pathologies in the images to the examining physician. This information assists doctors in quickly navigating through the collection of images, and it can help detect pathologies that may necessitate a more invasive diagnosis or immediate treatment. The technology owner is eager to engage in R&D collaboration with product development partners for gastro-intestinal endoscopic applications for both humans and animals.

Due to social stigma surrounding Sexually Transmitted Diseases (STD), cost barriers, and lack of awareness, a large proportion of men suffering from male health issues avoid accessing in-person medical care, instead, many men turn to online sources such as search engines or discussion forums to seek help. This further compounds the problem as this failure to seek medical attention leads to delays in diagnosis and subsequent treatment, while misinformation and misguidance from the general public can be dangerous and life-threatening. This technology aims to address the gap between crowd-sourced diagnosis and primary healthcare practitioners through a fully anonymous, AI-driven mobile application screening tool that covers 90% of genital pathologies e.g. bumps and lesions, and certain disease and viruses, such as syphilis, herpes, Human Papillomavirus (HPV). This technology offer is Artificial Intelligence (AI) enabled mobile application that utilises a Convolutional Neural Network (ConvNet or CNN) deep learning algorithm equipped with custom-built network layers to screen for male health issues, e.g. genital warts, sexually transmitted diseases (STD) etc. Data augmentation techniques and synthetic data generation methods have been used to vary the dataset and increase and sample size for realistic model training and testing. The highly accurate model has an accuracy of 60-90% for most cases of STI/STDs.

The latest Singapore National Population Health Survey has reported a concerning diabetes trend. From 2019-2020, 9.5% of the adults had diabetes, slightly dropping to 8.5% from 2021-2022. About 1 in 12 (8.5%) of residents aged 18 to 74 were diagnosed, with an age-standardised prevalence of 6.8% after accounting for population ageing. Among the diabetes patients, close to 1 in every 5 (18.8%) had undiagnosed diabetes, and 61.3% did not meet glucose control targets. Prediabetes is also prevalent, with 35% progressing to type 2 diabetes within eight years without lifestyle changes. Untreated Type 2 diabetes can lead to severe health issues. Tackling this challenge requires a holistic approach, focusing on awareness, early diagnosis, and lifestyle adjustments for diabetes and prediabetes. Recognising the need for innovation to address this, the technology owner develops a cost-effective and non-invasive AI-powered solution, Blood Glucose Evaluation And Monitoring (BGEM), that detects glucose dysregulation in individuals to monitor and evaluate diabetic risks. BGEM allows users to track their blood glucose levels regularly, identify any adverse trends and patterns, and adopt early intervention and lifestyle changes to prevent or delay the onset of diabetes. Clinically validated in 2022, with a research paper published in October 2023, the technology is open for licensing to senior care/home care providers, telehealth platforms, health wearables companies, and more.

There are 30,000 occupational drivers in Singapore, out of which 13,500 are 45 years old and above. The risk of acquiring cardiovascular disease increases with age and is potentially exacerbated by low physical activity and high emotional stress levels, which are two typical characteristics of occupational drivers arising from their work environment. Low level of physical activity and high stress levels have been shown to have significant relationship with heart rate variability, one of the indicators of cardiovascular disease. This technology is developed to help drivers to monitor their stress level, provide them with instantaneous feedback and the necessary alerts for a timely intervention. This technology offer presents a cross-platform AI system that estimates the stress levels continuously in real time, and can be easily integrated with commercially available photoplethysmography (PPG) wearables, e.g., a PPG wristwatch. In addition, this technology can be adapted for the monitoring of workplace stress with the aim of improving overall mental well-being.

Prolonged sitting has been identified as one of the risk factors associated with lower back pain (LBP). Slumped sitting posture, commonly observed in daily activities, results in spine flexion and increases the erector spinae (ES) muscle activity. This in turn elevates the joint compressive forces and the intra-muscular pressure, disrupting ES muscle blood flow and muscle oxygenation. Association has been established between muscle blood flow and fatigue. The near infrared spectroscopy (NIRS) is commonly used to measure muscle oxygenation. This non-invasive technology is commonly used by researchers due to its high accuracy. It is based on the absorption of near infrared light between the oxygenated and non-oxygenated hemoglobin and myoglobin. However, this is not well applicable in clinical settings due to its high costs and bulkiness.