Scientific Writings & Projects
CARTelligence: Artificial Intelligence-Driven Predictive Modelling of CAR-T Cell Therapy Outcomes in Cancer Treatment
A supervised machine learning framework developed using patient dataset for predicting patient outcomes to CAR-T therapy to help guide patient selection. Built using patient’s T cell-immunophenotyping data combined with clinical features, CAR-Telligence successfully captures the complex disease biology underlying CAR-T therapy.
​
Conferred Top 50 Asia S.T. Yau High School Award
​
Source code available here.
TUMOR MICROENVIRONMENT AND IMMUNOBIOLOGY: Data analysis
Contributed in analysis of data for the study that discovered that IL-9 also makes cancerous T cells move and spread more easily. When IL-9 activates its receptor, it creates a “pseudohypoxic” (low-oxygen-like) state inside the cells, increasing a protein called HIF-1α, which promotes cell movement.
​
Title of the Publication: IL-9 Promotes Migratory Dissemination of Malignant T Cells by Activating the HIF-1α–Cofilin-1 Axis in Cutaneous T-cell Lymphoma
​
Journal article available here.
Logistic Regression Model to Discern Disease Stage in Leukaemia (r/r B-ALL) Patients
Logistic regression model to accurately predict the stage of disease in r/r B-ALL patients using 2 biomarkers: Lactose Dehyrdogenase and extranodal spread.
​
This model was developed using a cohort of 60 patients diagnosed with r/r B-ALL at Tata Memorial Hospital, Mumbai.
​
Source code available here.
Cancer-Killing Viruses: The What, The How, and The What If
The convergence of oncolytic virotherapy and CAR-T cell technology heralds a new frontier in precision oncology. By leveraging the replicative lethality of viruses alongside the specificity of engineered lymphocytes, scientists are redefining what is possible in cancer immunotherapy. Though still in early clinical exploration, this combinatorial strategy offers a blueprint for durable, system-wide tumor eradication and could transform oncology in the coming decade.
​
Essay available here.
