Artificial Intelligence in Medicine: From Oncology to Predicting Preterm Birth

A new study by Harvard Medical School and Beth Israel Deaconess Medical Center has shown that in certain scenarios, modern artificial intelligence models outperform real doctors  in the accuracy of primary diagnosis. This is not yet a revolution in medicine, but it is already a significant shift toward the increasingly active use of AI. Read about the trends in the use of artificial intelligence in medicine in the UNN report. 

We should note immediately that this report is not about an "individual consultation" between a patient and ChatGPT or any other similar system — which we strongly urge you to refrain from — but about scientific developments that are gradually finding practical application in clinical settings. 

Oncology diagnostics - the largest field of application 

Last year, a group of researchers from the Korea Institute of Materials Science (KIMS) developed an optical biosensor capable of detecting trace amounts of cancer cell DNA in the blood with high sensitivity, enabling early cancer diagnosis. Researchers believe the sensor has higher sensitivity than a conventional biopsy. 

A study published in early 2026 states that Chinese systems DAMO GRAPE from Alibaba and RuiPath from Huawei have demonstrated high diagnostic accuracy and are already being used in clinical settings. It is noted that DAMO GRAPE for gastric cancer screening outperformed human radiologists, achieving a sensitivity of 85.1% and a specificity of 96.8%. In addition to detection, machine learning models are increasingly being used to predict treatment efficacy and patient outcomes.

Furthermore, the U.S. Food and Drug Administration (FDA) has approved over 70 AI-related devices, with 54.9% focused on radiology. Of these, the largest number of systems are aimed at general cancer detection, as well as breast cancer detection. AI-assisted technologies are also used during screenings, particularly for cervical cancer. Similar systems are gradually being introduced in Ukraine, specifically for the analysis of CT scans. 

AI and robotic medicine 

Another area where artificial intelligence is widely used is surgery. For example, systems can analyze a surgeon's movements during an operation and provide an analysis of the efficiency and improvement of their actions. Research shows that AI-assisted surgery reduces operation time by nearly a quarter. 

The most famous in the world is the Da Vinci robotic surgical system by Intuitive Surgical (USA), which allows a surgeon to perform minimally invasive operations through several small incisions (0.5–1 cm) instead of a large opening. Gradually, this system learns based on the acquired datasets and increases its efficiency. 

Perinatal medicine – from fetal heartbeat to predicting the due date 

Artificial intelligence in this field of medicine helps recognize anomalies in fetal development during ultrasound, and also identifies placenta previa during MRI scans. 

In March 2026, the American company Ultrasound AI received FDA registration for its Delivery Date AI technology. This technology demonstrates high accuracy in predicting the delivery date, relying exclusively on standard images taken during an ultrasound. Similar AI systems also help doctors analyze datasets and identify patterns that indicate, for example, the risk of excessive fetal weight. 

In Ukraine, as Dmytro Govseiev, Professor, Honored Doctor, and Head of the Department of Hospital Obstetrics and Gynecology and Postgraduate Education at the Bogomolets National Medical University, told UNN, AI-assisted studies are also beginning to be used, helping, in particular, to identify the risk of preterm labor. 

AI systems make it possible to predict the risk of preterm labor even with a minimal dataset. Combining traditional diagnostic methods with the latest models yields a very high result in forecasting. This gives doctors the time and opportunity to prolong the pregnancy and save the child's life 

According to him, AI is also quite actively used in cardiotocography — the study of the fetal heartbeat, which helps detect fetal distress (fetal discomfort caused, for example, by a lack of oxygen). 

At the same time, Dmytro Govseiev emphasizes that the ultimate responsibility for the patient's health lies with the doctor, and modern research systems are additional tools, not a replacement for a specialist.