Applied Intelligence Lead

Zina Ibrahim

Department of Biostatistics & Health Informatics

I will be updating this page with github links and relevant publications. For now, here's a synopsis of what I am up to.  All projects listed are open with active ongoing work. 

EQsPert

Extracting Qualitative Patterns from Electronic Records: Using principles of qualitative temporal reasoning, EQsPERt models longitudinal patient records as graphs of temporally-connected events. EQsPERt feeds these graphs as input to a Convolutional Neural Network to learn patterns of similarities between patients with similar phenotypes. The current proof-of-concept framework mines the Case Register Interactive Search (CRIS), a database containing the complete deanonymised health records of a large mental health hospital,  for patterns of comorbidities and drug-drug interactions from the records of psychiatric patients experiencing adverse drug reactions.

Agent Platform for automating patient PROvided Clinical outcome feedback: A distributed multi-agent system making up a virtual telehealth companion for patients after leaving the hospital. APPROaCh automates: 1) remote provision of Patient-reported Outcome Measures (PROMs), 2) automated assessment based on PROM values, 3) reminders for patients and intervention alerts to clinicians. APPROaCh also connects trialists by automating trial recruitment and collection of outcomes at end points. APPROaCh has been funded to conduct a trial evaluating its merit for patients with Attention Deficit Hyperactivity Disorder.

ADRIAN

Adverse Drug Reaction Intelligent AssistaNt: A rule-based multi-agent system for supporting clinical decisions to detect and prevent potential adverse drug reactions to prescribed medication. ADRIAn monitors the patient records for drug-drug interactions, drug-comorbidity risks and potential risk scores, consulting with clinical clinical guidelines to generate alerts and recommendations

Please reload