A Survey on Application of Visualization and AI Algorithm-Driven Technology for Healthcare

Healthcare facilities collect and produce vast amounts of clinical-relevant data. Various AI-related methods (like computer-aided detection for mammography and the learning and visualization of clinical pathways) are applied to healthcare these days, and visualization techniques are also used to support clinicians due to the complexities of clinical data. This self-contained survey focuses on the assessment of healthcare providers’ acceptance of and interaction with the AI algorithm-driven technology, along with the related visualization methods used in practice.
Historically, researchers have been concerned about the trust issue in computer-aided healthcare. For this reason,clinicians’ aversion and appreciation about algorithmic approaches in this field has been widely discussed. Some published studies have investigated physicians’ attitudes about technologies driven by AI algorithms, and generally, medical professionals seem to be somewhat skeptical about how well AI algorithms can perform in diagnostic tasks. Also, studies have been done on how healthcare providers indeed interact with algorithmic support systems in prognosis, diagnostics, and treatment recommendations, along with the visualization skills that are applied in practice.

Faculty Supervisor:

Marzyeh Ghassemi;Fanny Chevalier

Student:

Minfan Zhang

Partner:

Vector Institute

Discipline:

Computer science

Sector:

Professional, scientific and technical services

University:

University of Toronto