Using Machine Learning To Develop A Prediction Model As Companion Diagnostic To Guide The Use Of Chemoradiation In Esophageal Squamous Cell Carcinoma Patients
Grant Data
Project Title
Using Machine Learning To Develop A Prediction Model As Companion Diagnostic To Guide The Use Of Chemoradiation In Esophageal Squamous Cell Carcinoma Patients
Principal Investigator
Professor Law, Simon Ying Kit
(Principal Investigator (PI))
Duration
36
Start Date
2022-03-01
Amount
8446598
Conference Title
Using Machine Learning To Develop A Prediction Model As Companion Diagnostic To Guide The Use Of Chemoradiation In Esophageal Squamous Cell Carcinoma Patients
Keywords
Using Machine Learning , Develop A Prediction Model , Companion Diagnostic ,To Guide The Use Of Chemoradiation In Esophageal Squamous Cell Carcinoma Patients
Discipline
Surgical Research
HKU Project Code
MRP/025/20X
Grant Type
Midstream Research Programme for Universities (MRP)
Funding Year
2021
Status
On-going
Objectives
This R&D project advances precision oncology by developing a health technology for companion diagnostics. Our multidisciplinary team (biomedicine, computer science, basic/translational) will strategically work together based on our clinical expertise, data science knowledge, prior work/technical skills (filed patents, publications and preliminary data), patient resources (clinical database and surgical biobank of > 20 years) and long-term overseas collaboration. Importantly, we are in the best position in HK to implement our technology in a clinical setting. We will develop a first of its kind, clinically practicable, non-invasive, highly accurate, clinical/molecular data-driven prediction model for chemoradiation response for esophageal squamous cell carcinoma (ESCC) patients to personalize treatments. Currently, no clinical test for this purpose is globally accepted and used. Worldwide and in HK, ESCC is the most prevalent type of esophageal cancer that is clinically challenging and highly fatal. Locally advanced patients are treated with neoadjuvant chemoradiation and surgery. This trimodal therapy is effective in good responders by improving their outcomes and survivals, while poor responders may suffer unnecessary toxicities/side effects without clinical benefits. By removing unnecessary/ineffective treatments, our technology can reform the healthcare system to be more cost effective and improve the quality of life of patients.