Objective: Radiomics and deep transfer learning are two popular technologies used to develop computer-aided detection and diagnosis (CAD) schemes of medical images. This study aims to investigate and to compare the advantages and the potential limitations of applying these two technologies in developing CAD schemes. Methods: A relatively large and diverse retrospective dataset including 3000 digital mammograms was assembled in which 1496 images depicted malignant lesions and 1504 images depicted benign lesions. Two CAD schemes were developed to classify breast lesions. The first scheme was developed using four steps namely, applying an adaptive multi-layer topographic region growing algorithm to segment lesions, computing initial radiomics features, applying a principal component algorithm to generate an optimal feature vector, and building a support vector machine classifier. The second CAD scheme was built based on a pre-trained residual net architecture (ResNet50) as a transfer learning model to classify breast lesions. Both CAD schemes were trained and tested using a 10-fold cross-validation method. Several score fusion methods were also investigated to classify breast lesions. CAD performances were evaluated and compared by the areas under the ROC curve (AUC). Results: The ResNet50 model-based CAD scheme yielded AUC = 0.85 ± 0.02, which was significantly higher than the radiomics feature-based CAD scheme with AUC = 0.77 ± 0.02 (p < 0.01). Additionally, the fusion of classification scores generated by the two CAD schemes did not further improve classification performance. Conclusion: This study demonstrates that using deep transfer learning is more efficient to develop CAD schemes and it enables a higher lesion classification performance than CAD schemes developed using radiomics-based technology.

Objective: Since computer-aided diagnosis (CAD) schemes of medical images usually computes large number of image features, which creates a challenge of how to identify a small and optimal feature vector to build robust machine learning models, the objective of this study is to investigate feasibility of applying a random projection algorithm (RPA) to build an optimal feature vector from the initially CAD-generated large feature pool and improve performance of machine learning model. Methods: We assemble a retrospective dataset involving 1,487 cases of mammograms in which 644 cases have confirmed malignant mass lesions and 843 have benign lesions. A CAD scheme is first applied to segment mass regions and initially compute 181 features. Then, support vector machine (SVM) models embedded with several feature dimensionality reduction methods are built to predict likelihood of lesions being malignant. All SVM models are trained and tested using a leave-one-case-out cross-validation method. SVM generates a likelihood score of each segmented mass region depicting on one-view mammogram. By fusion of two scores of the same mass depicting on two-view mammograms, a case-based likelihood score is also evaluated. Results: Comparing with the principle component analyses, nonnegative matrix factorization, and Chi-squared methods, SVM embedded with RPA yielded a significantly higher case-based lesion classification performance with the area under ROC curve of 0.84 ± 0.01 (p<;0.02). Conclusion: The study demonstrates that RPA is a promising method to generate optimal feature vectors and improve SVM performance. Significance: This study presents a new method to develop CAD schemes with significantly higher and robust performance.

Both conventional and deep machine learning has been used to develop decision-support tools applied in medical imaging informatics. In order to take advantages of both conventional and deep learning approach, this study aims to investigate feasibility of applying a locally preserving projection (LPP) based feature regeneration algorithm to build a new machine learning classifier model to predict short-term breast cancer risk. First, a computer-aided image processing scheme was used to segment and quantify breast fibro-glandular tissue volume. Next, initially computed 44 image features related to the bilateral mammographic tissue density asymmetry were extracted. Then, an LLP-based feature combination method was applied to regenerate a new operational feature vector using a maximal variance approach. Last, a k-nearest neighborhood (KNN) algorithm based machine learning classifier using the LPP-generated new feature vectors was developed to predict breast cancer risk. A testing dataset involving negative mammograms acquired from 500 women was used. Among them, 250 were positive and 250 remained negative in the next subsequent mammography screening. Applying to this dataset, LLP-generated feature vector reduced the number of features from 44 to 4. Using a leave-onecase-out validation method, area under ROC curve produced by the KNN classifier significantly increased from 0.62 to 0.68 (p < 0.05) and odds ratio was 4.60 with a 95% confidence interval of [3.16, 6.70]. Study demonstrated that this new LPP-based feature regeneration approach enabled to produce an optimal feature vector and yield improved performance in assisting to predict risk of women having breast cancer detected in the next subsequent mammography screening.

In order to develop precision or personalized medicine, identifying new quantitative imaging markers and building machine learning models to predict cancer risk and prognosis has been attracting broad research interest recently. Most of these research approaches use the similar concepts of the conventional computer-aided detection schemes of medical images, which include steps in detecting and segmenting suspicious regions or tumors, followed by training machine learning models based on the fusion of multiple image features computed from the segmented regions or tumors. However, due to the heterogeneity and boundary fuzziness of the suspicious regions or tumors, segmenting subtle regions is often difficult and unreliable. Additionally, ignoring global and/or background parenchymal tissue characteristics may also be a limitation of the conventional approaches. In our recent studies, we investigated the feasibility of developing new computer-aided schemes implemented with the machine learning models that are trained by global image features to predict cancer risk and prognosis. We trained and tested several models using images obtained from full-field digital mammography, magnetic resonance imaging, and computed tomography of breast, lung, and ovarian cancers. Study results showed that many of these new models yielded higher performance than other approaches used in current clinical practice. Furthermore, the computed global image features also contain complementary information from the features computed from the segmented regions or tumors in predicting cancer prognosis. Therefore, the global image features can be used alone to develop new case-based prediction models or can be added to current tumor-based models to increase their discriminatory power.