Machine Learning and Radiomics in Nuclear Medicine and Molecular imaging: Part I. | ||||
| Egyptian Journal Nuclear Medicine | ||||
| Article 1, Volume 22, Issue 1, May and June 2021, Page 1-10 PDF (405.33 K) | ||||
| Document Type: Editorial | ||||
| DOI: 10.21608/egyjnm.2021.172714 | ||||
 View on SCiNiTO | ||||
| Author | ||||
Magdy M. Khalil
| ||||
| Medical Biophysics, Department of Physics, Faculty of Science, Helwan University. | ||||
| Abstract | ||||
| Diagnostic imaging modalities are undergoing a paradigm shift in terms of technological advances and their impact on many aspects of healthcare. The introduction of machine learning and radiomic in data analysis with capabilities of creating new clinical models has recently caught the attention of clinicians and scientists. Radiomics is a high throughput technology able to derive many imaging features from the diagnostic data while machine learning is a computer science discipline able to provide new forms of “electronic observer” able to mimic human tasks performed by radiologists and nuclear medicine physicians in daily routine. These technologies could be used individually or in combination to facilitate as well as solving issues associated with initial patient diagnosis, image processing, data analysis, stratification, prognosis and management. In the last decade, there was a rapidly growing interest in using radiomic in nuclear medicine and molecular imaging providing several solutions in reducing the injected radio activities, reducing imaging time, lesion segmentation, diagnosis, and many other applications that could potentially serve or replace current practices. The goal of this part of the machine learning and radiomics in nuclear medicine series is to introduce the reader to these new technologies and open avenues on current status, potential and future promises. | ||||
| Keywords | ||||
| Radiomics; Radio genomics; image segmentation; feature extraction; model validation | ||||
|
Statistics Article View: 231 PDF Download: 260 |
||||
