ICTS

" In recent years, the healthcare industry has shown increasing interest in developing inclusive and accessible artificial intelligence (AI) technologies. The objective is to ensure that AI-powered healthcare systems are accessible to everyone and inclusive of patients from diverse backgrounds. Ethical considerations such as privacy, security, and fairness also need to be taken into account during the design process. Head and Neck cancers (HNC) are classified as rare diseases, receiving less support compared to more prevalent cancers. However, HNC is a biologically complex and diverse group of rare malignancies, particularly burdensome in India. Diagnosis often occurs at advanced stages, resulting in poor prognosis and shorter lifespans. One of the major challenges after diagnosis is determining the most effective therapy for an individual tumor. Current decision-making relies on qualitative visual interpretations of radiological images, doctor-derived observations, and average statistics from clinical trials. However, advances in molecular biomarkers enable personalized cancer treatments based on specific genetic pathways. Radiological imaging remains crucial for diagnosis and decision-making, as it allows comprehensive analysis of the tumor. Despite these advancements, clinical outcomes in HNC remain unsatisfactory. Radiomics, a rapidly evolving field in oncology, uses AI-driven quantitative analysis of radiological imaging data. It converts qualitative images into numerical features for mining clinical insights using machine learning techniques. However, there are barriers to clinical adoption of radiomics in radiation oncology, including limited transparency, insufficient patient data for robust models, and a lack of randomized trials demonstrating added value. To address these limitations, our team at Christian Medical College, Vellore, India, funded by India Alliance DBT Wellcome, is working on better understanding imaging biomarkers in head and neck cancer through large-scale studies. We aim to overcome challenges and share our experiences with prospective imaging trials and multi-institutional radiomics studies. Integrating AI and quantitative imaging in healthcare has transformative potential, but addressing challenges is crucial for successful implementation and clinical impact."

In the UK alone, around 3000 babies (0.4 %) are tragically stillborn every year and nearly 60,000 (8 %) are born prematurely. Of those babies born early, some may not survive and others may face a lifetime of health issues. Evidence points to a wide geographical variation in the quality of care, attributable in part to disparities in how guidelines are followed and the availability of local resources.

Current pregnancy risk assessment is around 50 years old. It is based on paper/digital notes and asks only for the presence of certain risk factors. It does not investigate the interaction of those risk factors. It does not provide a numerical risk, thereby precluding personalisation of care. This results in poor triage into care pathways, inefficient use of limited medical resources and worsens health inequalities for underserved groups. Furthermore, despite digitisation of records, there is limited continuity of care information and no access to decision support. Digital notes do not present relevant information and, with over 90 NICE guidelines for maternity care, practitioners need context-specific decision support.

Our solution is the Tommy’s App, an innovative digital CE marked medical device which more accurately assesses a woman’s chance of preterm birth or of developing complications during pregnancy that can lead to stillbirth, and in doing so supports healthcare professionals to offer the right care at the right time and supports women to make informed decisions about their care.

The talk would encompass the varied pathologies afflicting the most visually critical part of the retina - the macula. The varied clinical manifestations of common, yet visually debilitating, macular pathologies such as age-related macular degeneration (ARMD), central serous chorioretinopathy (CSCR) and macular holes, a surgical condition, would be discussed. Insights regarding the underlying pathophysiology as well as imaging details would be shared too. Finally, a brief overview of AI predictive studies and models forecasting outcomes would be highlighted too.