Development of a Machine Learning-Based Model for Accurate Detection and Classification of Polycystic Ovary Syndrome on Pelvic Ultrasound.

Polycystic ovary syndrome (PCOS) is a common endocrine disorder that disrupts reproductive function and hormonal balance. It primarily affects reproductive-aged women and leads to physical, metabolic, and emotional challenges affecting the quality of life. In this study, we develop a machine learning-based model to accurately identify PCOS pelvic ultrasound images from normal pelvic ultrasound images. By leveraging 1,932 pelvic ultrasound images from the Kaggle online platform (Google LLC, Mountain View, CA), we were able to create a model that accurately detected multiple small follicles in the ovaries and an increase in ovarian volume for PCOS pelvic ultrasound images from normal pelvic ultrasound images. Our developed model demonstrated a promising performance, achieving a precision value of 82.6% and a recall value of 100%, including a sensitivity and specificity of 100% each. The value of the overall accuracy proved to be 100% and the F1 score was calculated to be 0.905. As the results garnered from our study are promising, further validation studies are necessary to generalize the model's capabilities and incorporate other diagnostic factors of PCOS such as physical exams and lab values.

Comparing Peer-Taught Student Tutors to Faculty-Taught Student Tutors in Educating Medical Students on Musculoskeletal Ultrasound.

INTRODUCTION: In recent years, medical education has witnessed a shift in the integration of ultrasound into the preclinical years of medical school. Given the exponential increase in accessibility to ultrasound technology, students now have the opportunity to create peer learning groups in which ultrasound concepts can be taught from peer to peer, empowering students to work together to integrate ultrasound concepts early in their preclinical education. This project investigates the efficacy of peer-taught student tutors (PTSTs) in imparting the fundamentals of basic ultrasound techniques to first-year medical students in the setting of identifying and labeling upper extremity musculoskeletal (MSK) anatomy.  Methods: First-year medical students were instructed to identify volar forearm structures with an ultrasound probe. Students and instructors were given access to an ultrasound probe, ultrasound gel, an iPad, and a standardized patient. Students were taught either by an ultrasound instructor (UI) or PTST. After a hands-on demonstration by a UI or PTST, participating students were told to take screenshots and label their images as accurately as possible, identifying the aforementioned volar structures on a standardized patient without any feedback. The labeled screenshot images of volar structures were graded based on the ability to clearly visualize the intended structures.  Results: The results of this study compare the efficacy of PTSTs as educators of basic sonographic identification techniques with that of UI faculty members. A chi-square analysis was performed between the images obtained by the UI and PTST students, and there was no statistically significant difference in identification accuracy between the groups (p = 0.7538, 0.1977, 0.1812, 0.301). When using the Mann-Whitney U rank test, there remained no statistically significant difference between the accuracy of the students taught by STs compared to students taught by UIs (p = 0.7744, 0.09538, 0.07547, 0.1846). Another finding showed that students belonging to both teaching groups were generally not able to infer the pathology of volar wrist structures when given pathology identification questions regarding upper extremity ultrasound. Using chi-square with Yates correction, there is no sufficient evidence to justify an association between the ability to answer pathology-based ultrasound questions and instructor type (p = p = 0.6299, 0.8725). CONCLUSIONS: This study supports the interpretation that the capability of first-year medical students to learn novice MSK sonographic identification is independent of whether the educator is a PTST or UI. This interpretation reveals a promising avenue toward the integration of the fundamentals of ultrasound identification early in medical education with little to no concern for the exhaustion of institutional resources. Along with the other well-documented benefits of the utilization of STs in medical school, a peer tutoring system centered on ultrasound skills designed in the way this study describes can be an effective, resource-sparing system that enhances medical students' sonographic capabilities early in their preclinical years.