New deep learning models lay the foundation for specialized diagnostic models

Researchers at Mount Sinai have developed an innovative artificial intelligence (AI) model for electrocardiogram (ECG) analysis that can interpret the electrocardiogram (ECG) as language. This approach can increase the accuracy and validity of ECG-related diagnoses, especially for cardiology where training data are limited.

In a study published online June 6, npj digital medicine, the team reported that a new deep learning model known as HeartBEiT forms the basis on which specialized diagnostic models can be created. The research team noted that the model created using HeartBEiT outperformed his established ECG analysis method in comparative tests.

Our model consistently outperformed convolutional neural networks [CNNs], is a machine learning algorithm commonly used for computer vision tasks. Such CNNs are often pre-trained on publicly available images of real-world objects.because Heartbeat Because it is ECG specific, it can perform as well or better than these methods using 1/10th of the data. This makes ECG-based diagnosis fairly viable, especially for rare conditions with small patient numbers and thus limited available data. “

Akhil Vaid, MD, First Author and Lecturer in Data-Driven and Digital Medicine Research, Mount Sinai Icahn School of Medicine

Over 100 million ECGs are performed each year in the United States alone, thanks to its low cost, non-invasiveness, and broad applicability in heart disease. Nonetheless, physicians are unable to consistently identify patterns that describe disease with the naked eye, particularly in situations where there are no established diagnostic criteria or where such patterns are subtle or chaotic for human interpretation. For conditions, the usefulness of the ECG is limited in scope. However, artificial intelligence is currently revolutionizing science, and most of the research so far has centered around his CNN.

Mount Sinai builds on his fascination with so-called generative AI systems such as ChatGPT to take the field in bold new directions. The system is built on Transformers, a deep learning model trained on huge datasets of texts to generate humans. Such as responding to user prompts on almost any topic. The researchers used a related image generation model to create a discrete representation of a small portion of his ECG, allowing the ECG to be analyzed as language.

“These representations can be considered individual words, and the entire ECG can be considered one document,” explains Dr. Vaid. “Heartbeat understands the relationships between these representations and uses this understanding to perform downstream diagnostic tasks more effectively. The three tasks we tested the model on were learning whether a patient was having a heart attack, whether they had a genetic condition called hypertrophic cardiomyopathy, and how efficiently their heart was functioning. was to do In both cases, our model outperformed all other baselines tested. “

pre-trained researchers Heartbeat It includes 8.5 million ECGs from 2.1 million patients collected over 40 years from four hospitals within the Mount Sinai medical system. We then tested its performance against standard CNN architectures for three cardiac diagnostic domains. As a result of research, Heartbeat The performance was significantly improved even with small sample sizes, and the ‘explainability’ was also improved. Senior Author Girish Nadkarni, M.D., MPH, Eileen and Arthur M. Fishberg, Ph.D., Icahn Mount Sinai Professor of Medicine, Director of the Charles Bronfman Institute for Personalized Medicine, Systems Director, Division of Data-Driven and Digital Medicine, Department of Medicine ) explained in detail. M.D.: “Neural networks are thought of as black boxes, but our model is more specific to the ECG regions involved in diagnosing heart attacks and other conditions, allowing clinicians to gain a deeper understanding of the underlying pathology. It helps to understand.CNN’s description was vague even when the diagnosis was pinpointed accurately.”

In fact, the Mount Sinai team has greatly enhanced the ways and opportunities a physician can interact with an ECG through a sophisticated new modeling architecture. “I want to make it clear that artificial intelligence will in no way replace electrocardiographic diagnosis by experts,” explained Dr. Nadkarni. “Rather, it augments the capabilities of artificial intelligence with exciting and compelling new ways to detect heart problems and monitor heart conditions,” Health. “