Although the use of artificial intelligence in image diagnosis is still in its early stages in Spain, compared to similar countries, artificial intelligence is the main topic of the congress of the Spanish Society of Medical Radiology (Seram), which will conclude today in Barcelona with the participation of about 2,400 specialists. Radiologists have not met in the Catalan capital for 20 years. No one imagined then that algorithms should contribute to diagnosis. All specialists are now convinced that AI will be used globally in various fields of medical care in the coming years, especially in radiology.
The revolution did not surprise professionals. “We are a clinical specialty, but it is also a technological specialty,” says Josep Munuera, Scientific Director of Ceram Hospital and Director of Radiodiagnosis at Sant Pau Hospital (Barcelona). “It is impossible not to talk about AI. We are the most advanced medical specialty in integrating AI into health and we have seen it at most tables and meetings these days,” he adds.
There is a different algorithm for each pathology; Incepto has developed 25 of them
80% of medical diagnoses are made through images. Artificial intelligence means applying algorithms and learning techniques to perform testing, interpret radiological images, and obtain images of the highest quality in less time. Finally, algorithms help identify the pathological condition and perform differential diagnosis. “We will demonstrate the value of AI in clinical practice by shortening the diagnosis time or being more accurate when classifying diseases,” says Dr. Munwira.
Current major applications of AI include bone fractures, especially in x-rays in the emergency department; Bone and soft tissue tumors on X-ray, CT, and MRI images; It helps in early detection and follow-up of joint and joint diseases, or in evaluating the condition of prosthetic limbs. Algorithms are even used to measure the amount of body fat for cardiovascular risk factors. All this is in the interpretive intelligence section, which is built on algorithms that facilitate reading tests. On the other hand, “non-interpretive intelligence is of great importance
Impact on experience
“Algorithms have been introduced that reduce the standard ring duration from 40 to 15 or 20 minutes, which maintains or improves image quality, with what that means for the patient who has to be inside the device,” says Munwira.
There is a different algorithm for each pathology. Incepto, which was present at the conference, has developed 25 devices. This French technology deals with breast cancer screening in Denmark, and lung tumor screening for the National Health Service in the United Kingdom, and its solutions are present in hundreds of European hospitals. Alfonso Martinez, the company’s general manager in Spain and Portugal, predicts that in the next five or ten years, diagnosis using artificial intelligence will be widely used because there is no doubt about the efficiency: “All Spanish hospitals will use it.”
In his opinion, the shortage of radiologists and the benefits of algorithms, along with the needs of patients, as they become increasingly older, are the factors that will drive this technology. Incepto trains the AI with thousands of images, and more than a million in some algorithms. Before marketing, a CE mark is required after the benefits provided by each algorithm have been proven through clinical trials. “In terms of breast cancer screening, we just published a report showing a 20% improvement in diagnosis and a 40% improvement in throughput.”
“Infuriatingly humble social media buff. Twitter advocate. Writer. Internet nerd.”
