ROME (ITALPRESS) – Artificial intelligence-based software has been developed that will be able to be incorporated into an ultrasound machine and ‘guide’the delivery, providing accurate, real-time information on the position of the baby’s head and then suggesting, clearly with a traffic light, to operators whether to proceed with natural descent into the birth canal, whether to use a suction cup or whether, even, it is better to switch to an emergency C-section. The tool, which could arrive in the delivery room starting in 2028, was developed and validated as part of work published in The European Journal of Obstetrics & Gynecology and Reproductive Biology and coordinated by Professor Tullio Ghi, since Nov. 1 full professor of Gynecology and Obstetrics at the Catholic University Rome campus, and Director of the Complex Operative Unit of Obstetrics at Policlinico Universitario Agostino Gemelli IRCCS, with technical support from the CNR Institute of Clinical Physiology in Lecce and colleagues at the Obstetric Clinic of the University of Parma. The multicenter study was carried out as part of the International Study Group on Labor and Delivery Sonography (ISLANDS) founded by Professor Ghi and will now continue with a new prospective research selected by the Ministry of Health among projects of national interest (PRIN) and funded with a fund of about 200 thousand euros.The fetus’ journey through the birth canal may not be without obstacles or risks: fetal malposition (an obstetrical condition in which the fetus’ occiput is oriented toward the sacrum and not toward the mother’s pubis) is among the most common causes of prolonged or arrested progression of delivery.Depending on how the baby’s head is positioned during descent then there may be a need to use a suction cup to facilitate exit or, in the most difficult cases, to resort to an emergency cesarean section for safe birth for mother and baby. Evaluating how the head is positioned may not be easy, even for the experienced operator, Professor Ghi explains: when the evaluation is done with the fingers, mistakes can be made in 1 in 5 cases, mistakes that can, for example, result in the application of the suction cup in the incorrect place on the head resulting in failure to extract the fetus , prolonging the delivery and, in the worst cases, causing a delay in the birth of a baby in distress .Ultrasound makes it easier for the doctor to accurately assess the position of the baby’s head before using the suction cup , but not all laborers working in the delivery room are able to make the best use of the ultrasound scanner to obtain accurate information . Artificial intelligence-based software uses ultrasound images to automatically give accurate answers in real time to the operators, who will display the ‘verdict’ as a traffic light, red if it is not appropriate to proceed with the suction cup and if the choice of emergency cesarean should be undertaken, green if the suction cup can be proceeded with, yellow if the situation is uncertain.In the published multicenter study, the software has been validated so far based on 2154 ultrasound images from 16 centers around the world. The overall performance of the model for classifying fetal head position was excellent, Professor Ghi explains, with an overall accuracy of 94.5 percent and a sensitivity of 95.6 percent (ability to detect head malposition). “We have developed an artificial intelligence model applied to ultrasound that can automatically and in a fraction of a second assess the position of the fetal head during delivery, with excellent overall accuracy,” Professor Ghi points out. “Future studies will need to validate our model on large patient populations before introducing it into routine clinical practice, but we believe that if the results are positive, the software could enter clinical practice within 3-4 years.
– photo press office Catholic University -(ITALPRESS).
