The increasing use of biomarkers in cancer have led to the concept of personalized medicine for patients. Personalized medicine provides better diagnosis and treatment options available to clinicians. Radiological imaging techniques provide an opportunity to deliver unique data on different types of tissue. However, obtaining useful information from all radiological data is challenging in the era of “big data”. Recent advances in computational power and the use of genomics have generated a new area of research termed Radiomics. Radiomics is defined as the high throughput extraction of quantitative imaging features or texture (radiomics) from imaging to decode tissue pathology and creating a high dimensional data set for feature extraction. Radiomic features provide information about the gray-scale patterns, inter-pixel relationships. In addition, shape and spectral properties can be extracted within the same regions of interest on radiological images. Moreover, these features can be further used to develop computational models using advanced machine learning algorithms that may serve as a tool for personalized diagnosis and treatment guidance.

Introduction: The radiological reading room is undergoing a paradigm shift to a symbiosis of computer science and radiology using artificial intelligence integrated with machine and deep learning with radiomics to better define tissue characteristics. The goal is to use integrated deep learning and radiomics with radiological parameters to produce a personalized diagnosis for a patient. Areas covered: This review provides an overview of historical and current deep learning and radiomics methods in the context of precision medicine in radiology. A literature search for ‘Deep Learning’, ‘Radiomics’, ‘Machine learning’, ‘Artificial Intelligence’, ‘Convolutional Neural Network’, ‘Generative Adversarial Network’, ‘Autoencoders’, Deep Belief Networks”, Reinforcement Learning”, and ‘Multiparametric MRI’ was performed in PubMed, ArXiv, Scopus, CVPR, SPIE, IEEE Xplore, and NIPS to identify articles of interest. Expert opinion: In conclusion, both deep learning and radiomics are two rapidly advancing technologies that will unite in the future to produce a single unified framework for clinical decision support with a potential to completely revolutionize the field of precision medicine.

Radiomics deals with the high throughput extraction of quantitative textural information from radiological images that not visually perceivable by radiologists. However, the biological correlation between radiomic features and different tissues of interest has not been established. To that end, we present the radiomic feature mapping framework to generate radiomic MRI texture image representations called the radiomic feature maps (RFM) and correlate the RFMs with quantitative texture values, breast tissue biology using quantitative MRI and classify benign from malignant tumors. We tested our radiomic feature mapping framework on a retrospective cohort of 124 patients (26 benign and 98 malignant) who underwent multiparametric breast MR imaging at 3 T. The MRI parameters used were T1-weighted imaging, T2-weighted imaging, dynamic contrast enhanced MRI (DCE-MRI) and diffusion weighted imaging (DWI). The RFMs were computed by convolving MRI images with statistical filters based on first order statistics and gray level co-occurrence matrix features. Malignant lesions demonstrated significantly higher entropy on both post contrast DCE-MRI (Benign-DCE entropy: 5.72 ± 0.12, Malignant-DCE entropy: 6.29 ± 0.06, p = 0.0002) and apparent diffusion coefficient (ADC) maps as compared to benign lesions (Benign-ADC entropy: 5.65 ± 0.15, Malignant ADC entropy: 6.20 ± 0.07, p = 0.002). There was no significant difference between glandular tissue entropy values in the two groups. Furthermore, the RFMs from DCE-MRI and DWI demonstrated significantly different RFM curves for benign and malignant lesions indicating their correlation to tumor vascular and cellular heterogeneity respectively. There were significant differences in the quantitative MRI metrics of ADC and perfusion. The multiview IsoSVM model classified benign and malignant breast tumors with sensitivity and specificity of 93 and 85%, respectively, with an AUC of 0.91.

Purpose: Treatment effect or radiation necrosis after stereotactic radiosurgery (SRS) for brain metastases is a common phenomenon often indistinguishable from true progression. Radiomics is an emerging field that promises to improve on conventional imaging. In this study, we sought to apply a radiomics-based prediction model to the problem of diagnosing treatment effect after SRS. Methods and Materials: We included patients in the Johns Hopkins Health System who were treated with SRS for brain metastases who subsequently underwent resection for symptomatic growth. We also included cases of likely treatment effect in which lesions grew but subsequently regressed spontaneously. Lesions were segmented semiautomatically on preoperative T1 postcontrast and T2 fluid-attenuated inversion recovery magnetic resonance imaging, and radiomic features were extracted with software developed in-house. Top-performing features on univariate logistic regression were entered into a hybrid feature selection/classification model, IsoSVM, with parameter optimization and further feature selection performed using leave-one-out cross-validation. Final model performance was assessed by 10-fold cross-validation with 100 repeats. All cases were independently reviewed by a board-certified neuroradiologist for comparison. Results: We identified 82 treated lesions across 66 patients, with 77 lesions having pathologic confirmation. There were 51 radiomic features extracted per segmented lesion on each magnetic resonance imaging sequence. An optimized IsoSVM classifier based on top-ranked radiomic features had sensitivity and specificity of 65.38% and 86.67%, respectively, with an area under the curve of 0.81 on leave-one-out cross-validation. Only 73% of cases were classifiable by the neuroradiologist, with a sensitivity of 97% and specificity of 19%. Conclusions: Radiomics holds promise for differentiating between treatment effect and true progression in brain metastases treated with SRS. A predictive model built on radiomic features from an institutional cohort performed well on cross-validation testing. These results warrant further validation in independent datasets. Such work could prove invaluable for guiding management of individual patients and assessing outcomes of novel interventions.