John F. Mccarthy, Md, Phd Email & Phone Number

Boston, Ma, Us

Louisville, Kentucky, Us

In my role as Associate Director for Humana's Digital Health and Analytics team, I was responsible for hiring and leading a group of data scientists in building models for predicting the risk of inpatient hospital admissions in our Medicare patient population using advanced analytics and ML/AI mehhodologies. These models were run on the Microsoft Azure Cloud using PySpark on Databricks. In my role as Lead Data Scientist, I was responsible for both the strategy and analytical aspects of two projects: 1) Identifying Modifiable Risk Factors in our Medicare population that could suggest interventions that might prevent hospitalization and 2) Identifying potential Social Determinants of Health (SDOH) factors that could potentially be addressed in reducing the risk of hospitalization. I also contributed both clinical and data science expertise in building three disease specific predictive models (cardiovascular, diabetes, and pulmonary) using combinations of both claims and clinical data.

Application of advanced analytics (ML and AI) to healthcare and life science data to extract new insights and transform them into actionable strategies for better decision making and improved outcomes.My interest in population health and preventive medicine motivated me to explore informatics based approaches to reducing the prevalence of obesity and its associated morbidities. This resulted in the development of a business plan and software prototype of a smartphone based SaaS platform for real-time decision support in behavioral weight management.I have driven both the business strategy and technical development aspects of several medical products and services. Recent engagements include an analytics platform for connected health and a clinical decision support platform for mHealth.

Somerville, Massachusetts, Us

I assisted in helping to define a new “value-based “ HEOR product offering for the payer/provider space. I also researched and explored the potential applicability of Bayesian causal modeling to type 2 diabetes using the GNS Healthcare proprietary REFS platform.

Boston, Ma, Us

I was responsible for contributing biostatistical expertise in analyzing clinical trial data in support of an FDA filing for a new drug-device combination (ITCA 650) for patients with type 2 diabetes. I was also the principal statistician and co-author on a publication comparing treatment satisfaction with ITCA 650 versus twice-daily injections of exenatide in type 2 diabetics using metformin.

New York, New York, Us

My position as an Associate Research Fellow and Lab Head in Computational Biology and Analytics at Pfizer involved building and leading a team of computational biologists in applying computational and systems biology approaches to the discovery of new drug targets and biomarker panels to improve the safety and efficacy of new drug candidates. My team was tasked with providing computational support for both the neuroscience and anti-infective therapeutic areas. In the neuroscience area, I led my team in using a combination of pathway analysis and machine learning methods to identify potential drug targets and understand the consequences of target modulation in bipolar disorder, schizophrenia, and Alzheimer’s disease. Our analysis resulted in more efficient utilization of experimental resources in addition to the identification of diagnostic disease biomarkers for which cost effective clinical assays could be developed.In the anti-infective area, I led my team in the development of a novel approach to theoretically predict bacterial cell surface antigens from primary sequence data, resulting in the identification of several new targets for vaccine development. In addition, I led the integration and analysis of gene expression, proteomic, metabolomic, and human clinical data to prioritize research efforts in identifying early diagnostic biomarkers for bacterial sepsis.

My involvement in the world of modeling and simulation in clinical trial data analysis was as a senior scientist and clinical consult for Pharsight Corporation. Here I combined my clinical knowledge with my biophysics and biostatistics background to construct and simulate pharmacokinetic and pharmacodynamic (PK/PD) models of clinical trial designs in order to maximize the probability of success. In addition to utilizing data from early clinical trials to optimize the design of later phase trials, I also built and simulated models constructed from data reported in clinical trial literature in order to improve strategic decision making throughout drug development and reduce overall cost. One such high impact project involved modeling and simulating the changes in lipid profile in response to various doses of standard lipid lowering therapies for comparison with a new chemical entity that was currently undergoing development.

My second commercial opportunity was with a VC funded start-up that focused on extracting biological knowledge from molecular and clinical databases using methods based on the application of machine learning and high dimensional visualization techniques.As Director of Discovery Informatics at AnVil, Inc., I built and led an interdisciplinary team in the successful mining of healthcare insurance claims databases and, in collaboration with the NCI, published an article on predictive modeling of non-small cell lung cancer on the basis of gene expression profiling. In addition, I first authored a separate paper discussing the application of machine learning and high dimensional visualization techniques in cancer research.

Us

My first non-academic position was with Cellomics, Inc. where I was hired to build and lead a new team in developing a web-based bioinformatics text mining tool for extracting and analyzing cellular pathway data derived through continuous mining of the National Library of Medicine (NLM) Medline database.By integrating my understanding of molecular biology with informatics knowledge derived from my academic training, I was able to lead a new team in delivering the first release of “CellSpace Knowledge Miner” software application for drug discovery research.

Pittsburgh, Pa, Us

After completing 3 years of training as a bench scientist, I decided to move my career away from data generation to focus more on data analysis and interpretation. This led to being awarded a 2 year National Library of Medicine fellowship to study biomedical informatics at the Center for Biomedical Informatics at the University of Pittsburgh.During my fellowship I received training in hospital information systems, machine learning, biostatistics, decision science, and knowledge representation which provided much of the foundation for future career opportunities.

Pittsburgh, Pennsylvania, Us

My interest in women's health and its genetic determinants led to an appointment as a research fellow at Magee-Womens Research Institute where I acquired training in molecular biology and molecular genetics as related to the pathophysiology of a high risk pregnancy complication known as preeclampsia. In addition to several genetic studies, I was the first to identify changes in maternal leptin concentrations in preeclampsia and suggest the possibility that alterations in maternal-fetal leptin regulation may play a significant role in the disease process. Subsequently, I was able to demonstrate that both leptin and its receptor were produced by placental tissues.

Pittsburgh, Pa, Us

Upon graduating from medical school I completed an internship in internal medicine at the University of Pittsburgh Medical Center which I chose due to its excellent reputation as an academic research center. My elective rotations in women's health and medical informatics during that year confirmed my desire to do research as opposed to patient care and set direction for my subsequent postdoctoral studies.

Phd, Biophysics

Mba, Strategy & Entrepreneurship

Mph, Epidemiology & Biostatistics

Md, Medicine (Current Massachusetts License)

Ms, Electrical Engineering | Biomedical Engineering

Ba, Physics And Chemistry