K20 : Pneumonia Detective: The Effectiveness of a Pneumonia Detection ML VS. Medical Practitioners


Students Zoya Waseem
School IND - Al-Falah Islamic School - Oakville
Level Junior 7/8 - Grade 8
Group Group 11 - Health Science IV
Abstract Pneumonia, a pulmonary infection impacting the alveoli in the lungs, resulted in 2.5 million fatalities in the year 2019. The current research initiative is designed to assess the efficacy of a machine learning model employing transfer learning techniques for pneumonia detection in contrast to the diagnostic capabilities of medical practitioners. The primary aim of this experiment is to investigate the potential of machine learning and transfer learning methodologies in identifying infiltrates in radiographic images. A positive outcome could position this tool as a valuable resource for validating diagnoses made by medical professionals, including physicians and radiologists, thereby mitigating the occurrence of misdiagnoses and subsequent delays in treatment. The implementation of such a tool has the potential to contribute to a reduction in mortality rates associated with pneumonia.
Awards
Group Award Prize
ArcelorMittal Dofasco AwardsArcelorMittal Dofasco Quality Systems Award$ 100
Hillfield Strathallan College Awards of ExcellenceHillfield Strathallan College Awards of Excellence - Life Sciences Award$ 100
Merit AwardsBronze Merit Award$ 60
McMaster University MGD Institute for Infectious Disease Research AwardsMichael G. DeGroote Institute for Infectious Disease Research Award$ 50