Multilayer Model for the Early Detection of Group A Streptococcus
Matan Porcilan, M.Sc. student at the department of Industrial
7 June 2022, 14:00 PM, Room 206& via zoom
Abstract:
Group A Streptococcus (GAS) is the most common bacterial cause of acute pharyngitis, accounting for ~30% of childhood cases and ~10% of adult cases. The current guideline for GAS diagnosis is based on the Centor clinical scoring system, which generates a score between -1 and +5 based on the age and several clinical symptoms of the patient. However, the current guideline suffers from high misdiagnosis rates of both types - patients with GAS who are sent home without treatment (20%), as well as patients without GAS who are prescribed antibiotics (35-50%). Here, we propose a multilayer model for the early detection of GAS, which provides a holistic view of the individual and the disease. The model incorporates data from multiple sources, including electronic medical records, smartphone and smartwatch sensors, self-reported questionnaires, and publicly available information. This allows combining information from multiple data sources, which are in principle independent, to increase the performance of the model. Further, if one or more of the layers are not available for a given individual, predictions can still be made. To evaluate our model, we analyzed data that we collected as part of the PerMed study. First, we analyzed the anonymized electronic medical records of 250,000 members of Maccabi Health Services, who were tested for 363,588 throat cultures in the last 12 years. In this setting, our model was able to obtain a considerably higher area under the ROC curve (AUC) compared with the existing Centor score (70% vs. 60% respectively). We also evaluated our model on a considerably richer but smaller cohort of 4,665 participants who were followed between April 2021 and March 2022 using our dedicated mobile application and smartwatches. These participants were tested 190 times for throat cultures during the study period. In this setting, the model was able to obtain a remarkably high AUC of 88%.
Bio:
Matan Porcilan is an M.Sc. student at the department of Industrial Engineering in Tel Aviv University, specializing in Data Science. Matan holds a B.Sc. degree in Industrial Engineering from Tel Aviv University. His research focuses on the early detection of Group A Streptococcus through the combined use of Electronic Medical Records, smartphones, smartwatches, and self-reported questionnaires. The research is being supervised by Dr. Erez Shmueli.
