Becker's Hospital Review

63 CIO / HEALTH IT For first time, AI beats experienced dermatologists in detecting skin cancer By Jessica Kim Cohen A n international team of researchers said they have developed an artificial intelligence system that diagnoses skin cancer more accurately than trained der- matologists. For the study, a team of researchers from Germany, the U.S. and France trained an AI system to identify skin cancer using 100,000- plus images, including images of malignant melanomas and benign moles. e research- ers used a type of AI known as a convolution- al neural network, which learns over time, rather than having to be programmed like typical soware. e researchers presented the trained convo- lutional neural network with 300 new images to determine whether the AI system was able to differentiate between cancerous and non- cancerous skin. e researchers found the convolutional neural network successfully identified 95 percent of the melanomas and 71.3 percent of the benign moles, according to results published in the Annals of Oncology. To assess the outcomes of the convolutional neural network, the researchers also enrolled 58 dermatologists from 17 countries to eval- uate 100 of the cases. When provided with clinical information and close-up images of the cases, the dermatologists were able to ac- curately diagnose 88.9 percent of malignant melanomas and 75.7 percent of benign moles. e researchers concluded the convolutional neural network missed fewer melanomas and misdiagnosed benign moles less frequently than the dermatologists. is study marks the first time a form of AI has reported better results at detecting melanomas than trained dermatologists, according to a May 28 state- ment from the European Society for Medical Oncology. "ese findings show that deep learning convolutional neural networks are capable of out-performing dermatologists, includ- ing extensively trained experts, in the task of detecting melanomas," Holger Haenssle, a senior managing physician in the dermatolo- gy department at University of Heidelberg in Germany and first author of the study, said in the statement. n Intermountain expands precision genomics DNA test to all providers By Jessica Kim Cohen P roviders across Intermountain Healthcare now have access to RxMatch, the Salt Lake City-based health system's precision genomics service. Through the program, a physician is able to collect a patient's DNA sample using a cheek swab. The Rx- Match Comprehensive Panel then analyzes 97 genes from the DNA sample to determine how a patient will likely respond to various medications, including opioids, statins, immunosuppressants and antide- pressants. Physicians can use the panel's results to guide appropriate dosage and prescriptions. The program falls under the umbrella of pharma- cogenomics, or the study of how genetics affect a patient's response to medication. "The objective of this project is to provide the most comprehensive and evidence-based information to the physician, thus decreasing the amount of time and money spent to achieve the correct medication," David Loughmiller, laboratory manager for Inter- mountain Precision Genomics, said in a May 8 state- ment. n Global telemedicine market to hit $48.8B by 2023 By Anuja Vaidya T he global telemedicine market is expected to experience rapid growth over the next five years, according to a P&S Market Research report. Here are five key trends: 1. The market is anticipated to reach $48.8 billion by 2023. 2. Analysts predict it will grow at a compound annual growth rate of 14.8 percent between 2018 and 2023. 3. Major drivers of the market include: • The growing geriatric population • The increased need for remote patient monitoring services • The growing prevalence of chronic and lifestyle-associated diseases 4. However, inadequate healthcare structures in developing economies are proving to be a barrier to the growth of the glob- al telemedicine market. 5. The market is seeing several collaborations and partner- ships focused on accelerating and streamlining solutions as well as providing a wide range of services to healthcare or- ganizations. n

• Cover