A5004615169- Health Systems, Economic Evaluations, Quality of Life
- Digital Mental Health Interventions
- Advanced Causal Inference Techniques
- Electronic Health Records Systems
- Artificial Intelligence in Healthcare and Education
- Machine Learning and Data Classification
- Telemedicine and Telehealth Implementation
- Statistical Methods in Clinical Trials
- Health disparities and outcomes
- Chronic Disease Management Strategies
- Advanced Bandit Algorithms Research
- Schizophrenia research and treatment
- Mobile Health and mHealth Applications
- Anomaly Detection Techniques and Applications
- Machine Learning in Healthcare
- Machine Learning and Algorithms
- Mental Health Research Topics
- Reinforcement Learning in Robotics
- Technology Use by Older Adults
- COVID-19 and Mental Health
- Global Cancer Incidence and Screening
- Artificial Intelligence in Healthcare
- Sex and Gender in Healthcare
- Advanced Multi-Objective Optimization Algorithms
- Medical Coding and Health Information
Western University
2016-2025
Access Alliance Multicultural Health and Community Services
2023
Digital Research Alliance of Canada
2023
Western University of Health Sciences
2023
Lawson Health Research Institute
2022-2023
National Center for Health Statistics
2022
University of Waterloo
2011-2021
St. Michael's Hospital
2016
University of Alberta
2002-2012
University of Michigan
2010-2011
Abstract The selection of effective genes that accurately predict chemotherapy responses might improve cancer outcomes. We compare optimized gene signatures for cisplatin, carboplatin, and oxaliplatin in the same cell lines validate each signature using data from patients with cancer. Supervised support vector machine learning is used to derive sets whose expression related line GI 50 values by backwards feature cross-validation. Specific functional pathways distinguishing sensitive...
Dynamic treatment regimes are of growing interest across the clinical sciences because these provide one way to operationalize and thus inform sequential personalized decision making. Formally, a dynamic regime is sequence rules, per stage intervention. Each rule maps up-to-date patient information recommended treatment. We briefly review variety approaches for using data construct rules. then critical inferential challenge that results from nonregularity, which often arises in this area. In...
Intersectionality recognizes that in the context of sociohistorically shaped structural power relations, an individual's multiple social positions or identities (e.g., gender, ethnicity) can interact to affect health-related outcomes. Despite limited methodological guidance, intersectionality frameworks have increasingly been incorporated into epidemiological studies, both describe health disparities and examine their causes. This study aimed advance methods intersectional estimation binary...
Intersectionality theoretical frameworks have been increasingly incorporated into quantitative research. A range of methods applied to describing outcomes and disparities across large numbers intersections social identities or positions, with limited evaluation.Using data simulated reflect plausible epidemiologic scenarios, we evaluated for intercategorical intersectional analysis continuous outcomes, including cross-classification, regression interactions, multilevel individual...
Abstract Background Effective deployment of AI tools in primary health care requires the engagement practitioners development and testing these tools, a match between resulting clinical/system needs care. To set stage for developments, we must gain more in-depth understanding views decision-makers about use The objective this study was to identify key issues regarding by exploring digital stakeholders. Methods This utilized descriptive qualitative approach, including thematic data analysis....
We present an efficient "sparse sampling" technique for approximating Bayes optimal decision making in reinforcement learning, addressing the well known exploration versus exploitation tradeoff. Our approach combines sparse sampling with Bayesian to achieve improved while controlling computational cost. The idea is grow a lookahead tree, intelligently, by exploiting information posterior---rather than enumerate action branches (standard sampling) or compensate myopically (value of perfect...
Summary Dynamic treatment regimes (DTRs) operationalize the clinical decision process as a sequence of functions, one for each decision, where function maps up‐to‐date patient information to single recommended treatment. Current methods estimating optimal DTRs, example Q ‐learning, require specification outcome by which “goodness” competing dynamic is measured. However, this an over‐simplification goal making, aims balance several potentially outcomes, example, symptom relief and side‐effect...
<h3>Purpose:</h3> To understand staff and health care providers' views on potential use of artificial intelligence (AI)-driven tools to help for patients within a primary setting. <h3>Methods:</h3> We conducted qualitative descriptive study using individual semistructured interviews. As part province-wide Learning Health Organization, Community Centres (CHCs) are community-governed, team-based delivery model providing people who experience marginalization in Ontario, Canada. CHC providers...
The Ontario electrical grid is sized to meet peak electricity load. A reduction in load would allow deferring large infrastructural costs of additional power plants, thereby lowering generation cost and prices. Proposed solutions for include demand response storage. Both these require accurate prediction a home's mean Existing work has focused only on prediction. We find that methods exhibit high error when predicting Moreover, historic occupancy better predictor than observable physical...
Internationally, primary care practice had to transform in response the COVID pandemic. Informatics issues included access, privacy, and security, as well patient concerns of equity, safety, quality, trust. This paper describes progress lessons learned.IMIA Primary Care Working Group members from Australia, Canada, United Kingdom States developed a standardised template for collection information. The guided rapid literature review. We also experiential learning public health...
Our increasingly connected world continues to face an ever-growing number of network-based attacks. An Intrusion Detection System (IDS) is essential security technology used for detecting these Although numerous Machine Learning-based IDSs have been proposed the detection malicious network traffic, majority difficulty properly and classifying more uncommon attack types. In this paper, we implement a novel hybrid technique using synthetic data produced by Generative Adversarial Network (GAN)...
Despite widespread advancements in and envisioned uses for artificial intelligence (AI), few examples of successfully implemented AI innovations exist primary care (PC) settings.To identify priority areas PC Ontario, Canada.A collaborative consultation event engaged multiple stakeholders a nominal group technique process to generate, discuss rank ideas how can support Ontario PC.The produced nine ranked priorities: (1) preventative risk profiling, (2) patient self-management condition(s),...
There is a widespread and longstanding belief that machine learning models are biased towards the majority (or negative) class when from imbalanced data, leading them to neglect or ignore minority positive) class. In this study, we show not necessarily correct for decision trees, their bias can actually be in opposite direction. Motivated by recent simulation study suggested trees class, our paper aims reconcile conflict between decades of other works. First, critically evaluate past...