Radiomics and Deep Learning Image

Radiomics and Deep Learning Advances Have Potential to Transform Precision Medicine

By HealthMyne | July 23, 2019 | Blog

Analysis by two Johns Hopkins scholars suggests the simultaneous advance of two emerging technologies – deep learning and radiomics – will empower a single, unified framework for clinical decision support with potential to revolutionize the field of precision medicine.

Presented in the journal Expert Review of Precision Medicine and Drug Development, the analysis predicts deep learning and radiomic methods will transform medical imaging and its application to personalized medicine, allowing rapid mining of patient data and radiological imaging biomarkers.

“These findings support our belief that in the near future radiomics will completely change the way medicine is practiced,” says Neal Miller, HealthMyne’s vice president of marketing. “We’re especially excited about the power of radiomics in oncology. A growing body of evidence points to the promise of radiomics as a methodology with strong predictive capabilities in diagnosis, prognosis and prescription. One application is ‘virtual biopsy’ in which the image alone can tell you whether someone has cancer or not and even what type of cancer it is. Imagine how much time, money and pain and suffering could be saved as this and other uses of radiomics become mainstream.”

The journal article defines radiomics as quantitative measures of textures in radiological medical images at the micro-level, where aspects like interpixel relationships and spectral properties can be teased out mathematically. Precision medicine, aka “personalized” medicine, is described as an approach to disease treatment and prevention that incorporates data on genetics, environment and lifestyle at the level of the individual patient.

Miller says the HealthMyne Platform creates 1500+ metrics for every cancer lesion identified and saves that data to a discoverable database. From simple evidence-based metrics to more complex radiomics such as texture and spiculation, data are easily mineable and available for inclusion in precision medicine initiatives and for clinical, translational, and commercial research.

For more information on HealthMyne’s application of radiomics in oncology, email us at [email protected].