Early detection of obstetric anal sphincter injuries (OASI) presents a significant challenge due to lack of both effective and practically available diagnostic tools. Thus, ONIRY, a medical device employing electrical impedance spectroscopy combined with machine learning (ML), was introduced to detect OASI quickly after vaginal delivery. For this post-hoc analysis of the ML approach for determining the OASI location, data was utilized from a clinical study that enrolled 152 women post-vaginal delivery with varying degrees of perineal injuries or no such injuries. Both endoanal ultrasound (EUS), as the reference method, and impedance spectroscopy were performed. ML classification was also performed. The ML model developed within this analysis not only detects OASI but also suggests the location the injury detected. This location analysis had an average accuracy of 85.6% compared with the location indicated by investigators based on EUS.
Determination of anal sphincter injury location using impedance spectroscopy in obstetric patients / Mlynczak, M.; Rosol, M.; Korzeniewski, K.; Iwanowski, P.; Salvatore, S.; Ratto, C.; Spinelli, A.; Borycka, K.. - (2024). ( 46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2024 Disney�s Coronado Springs Convention Center, usa 2024) [10.1109/EMBC53108.2024.10782955].
Determination of anal sphincter injury location using impedance spectroscopy in obstetric patients
Salvatore S.;
2024-01-01
Abstract
Early detection of obstetric anal sphincter injuries (OASI) presents a significant challenge due to lack of both effective and practically available diagnostic tools. Thus, ONIRY, a medical device employing electrical impedance spectroscopy combined with machine learning (ML), was introduced to detect OASI quickly after vaginal delivery. For this post-hoc analysis of the ML approach for determining the OASI location, data was utilized from a clinical study that enrolled 152 women post-vaginal delivery with varying degrees of perineal injuries or no such injuries. Both endoanal ultrasound (EUS), as the reference method, and impedance spectroscopy were performed. ML classification was also performed. The ML model developed within this analysis not only detects OASI but also suggests the location the injury detected. This location analysis had an average accuracy of 85.6% compared with the location indicated by investigators based on EUS.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
