Comprehensive Summary
Rachel Chacko et al. investigated how the implementation of deep learning basal cell carcinoma (BCC) detection and tumor mapping technology affected the working efficiency of clinical and laboratory staff. The study was performed on 104 consecutive Mohs micrographic surgeries (MMS) over a 20-day period. This AI technology was intended to provide intraoperative tissue grossing and inking recommendations, make assessments on tissue section quality, and map the neoplasm to aid in the removal process. Three measures were defined that evaluated algorithm integration: slide waiting time, staff waiting time, and histotechnician waiting time. The results of the simulated study showed that the deep learning algorithm led to 35.6% reduction in slide waiting time, 18.6% reduction histotechnician waiting time, and 18.4% reduction staff waiting time per day. Although the AI algorithm was successful, the authors discussed concerns of computing power, which may not be typical for a regular hospital.
Outcomes and Implications
As highlighted in the study, over 195 thousand additional registered nurses and 112 thousand medical assistants are estimated to be necessary by 2031 to care for the aging populations. With this large demand, the implementation of AI can alleviate this stress by automating protocols that would otherwise be performed by humans. With the automation of these processes, healthcare workers can place their attention on other patient-related needs, therby enhancing care. Given the significant reduction in wait times seen via the implementation of this deep learning algorithm, the usage of this technology is recommended. In order to facilitate this technology’s expansion in the clinical sector, addressing the computing power required to run this deep learning algorithm is critical. Overcoming such barriers will facilitate the growth of this program and can enhance patient outcomes.