Left ventricular diastolic dyfunction detection is particularly important in cardiac function screening. This paper proposed a phonocardiogram (PCG) transfer learning-based CatBoost model to detect diastolic dysfunction noninvasively. The Short-Time Fourier Transform (STFT), Mel Frequency Cepstral Coefficients (MFCCs), S-transform and gammatonegram were utilized to perform four different representations of spectrograms for learning the representative patterns of PCG signals in two-dimensional image modality. Then, four pre-trained convolutional neural networks (CNNs) such as VGG16, Xception, ResNet50 and InceptionResNetv2 were employed to extract multiple domain-specific deep features from PCG spectrograms using transfer learning, respectively. Further, principal component analysis and linear discriminant analysis (LDA) were applied to different feature subsets, respectively, and then these different selected features are fused and fed into CatBoost for classification and performance comparison. Finally, three typical machine learning classifiers such as multilayer perceptron, support vector machine and random forest were employed to compared with CatBoost. The hyperparameter optimization of the investigated models was determined through grid search. The visualized result of the global feature importance showed that deep features extracted from gammatonegram by ResNet50 contributed most to classification. Overall, the proposed multiple domain-specific feature fusion based CatBoost model with LDA achieved the best performance with an area under the curve of 0.911, accuracy of 0.882, sensitivity of 0.821, specificity of 0.927, F1-score of 0.892 on the testing set. The PCG transfer learning-based model developed in this study could aid in diastolic dysfunction detection and could contribute to non-invasive evaluation of diastolic function.
Purpose: To investigate the predictive performance of machine learning-based CT radiomics for differentiating between low- and high-nuclear grade of clear cell renal cell carcinomas (CCRCCs).Methods: This retrospective study enrolled 406 patients with pathologically confirmed low- and high-nuclear grade of CCRCCs according to the WHO/ISUP grading system, which were divided into the training and testing cohorts. Radiomics features were extracted from nephrographic-phase CT images using PyRadiomics. A support vector machine (SVM) combined with three feature selection algorithms such as least absolute shrinkage and selection operator (LASSO), recursive feature elimination (RFE), and ReliefF was performed to determine the most suitable classification model, respectively. Clinicoradiological, radiomics, and combined models were constructed using the radiological and clinical characteristics with significant differences between the groups, selected radiomics features, and a combination of both, respectively. Model performance was evaluated by receiver operating characteristic (ROC) curve, calibration curve, and decision curve analyses.Results: SVM-ReliefF algorithm outperformed SVM-LASSO and SVM-RFE in distinguishing low- from high-grade CCRCCs. The combined model showed better prediction performance than the clinicoradiological and radiomics models (p < 0.05, DeLong test), which achieved the highest efficacy, with an area under the ROC curve (AUC) value of 0.887 (95% confidence interval [CI] 0.798–0.952), 0.859 (95% CI 0.748–0.935), and 0.828 (95% CI 0.731–0.929) in the training, validation, and testing cohorts, respectively. The calibration and decision curves also indicated the favorable performance of the combined model.Conclusion: A combined model incorporating the radiomics features and clinicoradiological characteristics can better predict the WHO/ISUP nuclear grade of CCRCC preoperatively, thus providing effective and noninvasive assessment.
Background and Objective: Vascular dementia (VaD) and Alzheimer's disease (AD) could be characterized by the same syndrome of dementia. This study aims to assess whether multi-parameter features derived from structural MRI can serve as the informative biomarker for differential diagnosis between VaD and AD using machine learning. Methods: A total of 93 patients imaged with brain MRI including 58 AD and 35 VaD confirmed by two chief physicians were recruited in this study from June 2013 to July 2019. Automated brain tissue segmentation was performed by the AccuBrain tool to extract multi-parameter volumetric measurements from different brain regions. Firstly, a total of 62 structural MRI biomarkers were addressed to select significantly different features between VaD and AD for dimensionality reduction. Then, the least absolute shrinkage and selection operator (LASSO) was further used to construct a feature set that is fed into a support vector machine (SVM) classifier. To ensure the unbiased evaluation of model performance, a comparative study of classification models was implemented by using different machine learning algorithms in order to determine which performs best in the application of differential diagnosis between VaD and AD. The diagnostic performance of the classification models was evaluated by the quantitative metrics derived from the receiver operating characteristic curve (ROC). Results: The experimental results demonstrate that the SVM with RBF achieved an encouraging performance with sensitivity (SEN), specificity (SPE), and accuracy (ACC) values of 82.65%, 87.17%, and 84.35%, respectively (AUC = 0.861, 95% CI = 0.820–0.902), for the differential diagnosis between VaD and AD. Conclusions: The proposed computer-aided diagnosis method highlights the potential of combining structural MRI and machine learning to support clinical decision making in distinction of VaD vs. AD.
Background: The computed tomography (CT) features of small solid lung cancers and their changing regularity as they grow have not been well studied. The purpose of this study was to analyze the CT features of solid lung cancerous nodules (SLCNs) with different sizes and their variations. Methods: Between February 2013 and April 2018, a consecutive cohort of 224 patients (225 nodules) with confirmed primary SLCNs was enrolled. The nodules were divided into four groups based on tumor diameter (A: diameter ≤ 1.0 cm, 35 lesions; B: 1.0 cm < diameter ≤ 1.5 cm, 60 lesions; C: 1.5 cm < diameter ≤ 2.0 cm, 63 lesions; and D: 2.0 cm < diameter ≤ 3.0 cm, 67 lesions). CT features of nodules within each group were summarized and compared. Results: Most nodules in different groups were located in upper lobes (groups A − D:50.8%–73.1%) and had a gap from the pleura (groups A − D:89.6%–100%). The main CT features of smaller (diameter ≤ 1 cm) and larger (diameter > 1 cm) nodules were significantly different. As nodule diameter increased, more lesions showed a regular shape, homogeneous density, clear but coarse tumor–lung interface, lobulation, spiculation, spinous protuberance, vascular convergence, pleural retraction, bronchial truncation, and beam-shaped opacity (p < 0.05 for all). The presence of halo sign in all groups was similar (17.5%–22.5%; p > 0.05). Conclusions: The CT features vary among SLCNs with different sizes. Understanding their changing regularity is helpful for identifying smaller suspicious malignant nodules and early determining their nature in follow-up.