no code implementations • 29 Apr 2025 • Lovedeep Gondara, Jonathan Simkin, Graham Sayle, Shebnum Devji, Gregory Arbour, Raymond Ng
This study aims to guide language model selection by investigating: 1) the necessity of finetuning versus zero-shot usage, 2) the benefits of domain-adjacent versus generic pretrained models, 3) the value of further domain-specific pretraining, and 4) the continued relevance of Small Language Models (SLMs) compared to Large Language Models (LLMs) for specific tasks.
no code implementations • 21 Apr 2025 • Mohammad Beheshti, Lovedeep Gondara, Iris Zachary
Discussion: Fine-tuned language models achieved strong performance in patient record blocking and matching with minimal errors.
no code implementations • 24 Mar 2025 • Lovedeep Gondara, Jonathan Simkin, Shebnum Devji, Gregory Arbour, Raymond Ng
Population-based cancer registries (PBCRs) face a significant bottleneck in manually extracting data from unstructured pathology reports, a process crucial for tasks like tumor group assignment, which can consume 900 person-hours for approximately 100, 000 reports.
no code implementations • 11 Nov 2024 • Lovedeep Gondara, Jonathan Simkin
We present an audit mechanism for language models, with a focus on models deployed in the healthcare setting.
1 code implementation • 9 Dec 2021 • Lovedeep Gondara, Ke Wang, Ricardo Silva Carvalho
Learning from continuous data streams via classification/regression is prevalent in many domains.
1 code implementation • 16 Feb 2020 • Lovedeep Gondara, Ke Wang, Ricardo Silva Carvalho
We propose the differentially private lottery ticket mechanism (DPLTM).
no code implementations • 4 Oct 2019 • Lovedeep Gondara, Ke Wang
Survival function estimation is used in many disciplines, but it is most common in medical analytics in the form of the Kaplan-Meier estimator.
no code implementations • 25 Sep 2019 • Lovedeep Gondara, Ke Wang, Ricardo Silva Carvalho
We propose the differentially private lottery ticket mechanism (DPLTM).
no code implementations • 12 Feb 2018 • Lovedeep Gondara, Ke Wang
Loss to followup is a significant issue in healthcare and has serious consequences for a study's validity and cost.
3 code implementations • 8 May 2017 • Lovedeep Gondara, Ke Wang
Missing data is a significant problem impacting all domains.
no code implementations • 5 May 2017 • Lovedeep Gondara
Machine learning models, especially based on deep architectures are used in everyday applications ranging from self driving cars to medical diagnostics.
no code implementations • 8 Dec 2016 • Lovedeep Gondara
Can humans impute missing data with similar proficiency as machines?
no code implementations • 29 Sep 2016 • Lovedeep Gondara
New proposed models are often compared to state-of-the-art using statistical significance testing.
2 code implementations • 16 Aug 2016 • Lovedeep Gondara
Image denoising is an important pre-processing step in medical image analysis.