The purpose of the present study was to examine the potential of a machine learning model with integrated clinical and CT-based radiomics features in predicting pathologic complete response (pCR) to neoadjuvant systemic therapy (NST) in breast cancer. Contrast-enhanced CT was performed in 329 patients with breast tumors (n = 331) before NST. Pyradiomics was used for feature extraction, and 107 features of seven classes were extracted. Feature selection was performed on the basis of the intraclass correlation coefficient (ICC), and six ICC thresholds (0.7–0.95) were examined to identify the feature set resulting in optimal model performance. Clinical factors, such as age, clinical stage, cancer cell type, and cell surface receptors, were used for prediction. We tried six machine learning algorithms, and clinical, radiomics, and clinical–radiomics models were trained for each algorithm. Radiomics and clinical–radiomics models with gray level co-occurrence matrix (GLCM) features only were also built for comparison. The linear support vector machine (SVM) regression model trained with radiomics features of ICC ≥0.85 in combination with clinical factors performed the best (AUC = 0.87). The performance of the clinical and radiomics linear SVM models showed statistically significant difference after correction for multiple comparisons (AUC = 0.69 vs. 0.78; p < 0.001). The AUC of the radiomics model trained with GLCM features was significantly lower than that of the radiomics model trained with all seven classes of radiomics features (AUC = 0.85 vs. 0.87; p = 0.011). Integration of clinical and CT-based radiomics features was helpful in the pretreatment prediction of pCR to NST in breast cancer.

Anode heel effects are known to cause non-uniform image quality, but no method has been proposed to evaluate the non-uniform image quality caused by the heel effect. Therefore, the purpose of this study was to evaluate non-uniform image quality in digital radiographs using a novel circular step-wedge (CSW) phantom and normalized mutual information (nMI). All X-ray images were acquired from a digital radiography system equipped with a CsI flat panel detector. A new acrylic CSW phantom was imaged ten times at various kVp and mAs to evaluate overall and non-uniform image quality with nMI metrics. For comparisons, a conventional contrast-detail resolution phantom was imaged ten times at identical exposure parameters to evaluate overall image quality with visible ratio (VR) metrics, and the phantom was placed in different orientations to assess non-uniform image quality. In addition, heel effect correction (HEC) was executed to elucidate the impact of its effect on image quality. The results showed that both nMI and VR metrics significantly changed with kVp and mAs, and had a significant positive correlation. The positive correlation is suggestive that the nMI metrics have a similar performance to the VR metrics in assessing the overall image quality of digital radiographs. The nMI metrics significantly changed with orientations and also significantly increased after HEC in the anode direction. However, the VR metrics did not change significantly with orientations or with HEC. The results indicate that the nMI metrics were more sensitive than the VR metrics with regards to non-uniform image quality caused by the anode heel effect. In conclusion, the proposed nMI metrics with a CSW phantom outperformed the conventional VR metrics in detecting non-uniform image quality caused by the heel effect, and thus are suitable for quantitatively evaluating non-uniform image quality in digital radiographs with and without HEC.

Background: Trigeminal neuralgia (TN) is the most common type of chronic neuropathic facial pain, but the etiology and pathophysiological mechanisms after treatment are still not well understood. The purpose of this study was to investigate the longitudinal changes of the cisternal segment of the trigeminal nerve and brain pain-related regions in patients with TN before and after treatment using readout segmentation of long variable echo-train (RESOLVE) diffusion tensor imaging (DTI) and transverse relaxation (T2)-weighted sampling perfection with application-optimized contrast at different flip angle evolutions (T2-SPACE). Methods: Twelve patients with TN and four healthy controls were enrolled in this study. All patients underwent assessment of the visual analog scale (VAS), and acquisition of RESOLVE DTI and T2-SPACE images before and at 1, 6, and 12 months after treatments. Regions-of-interest were placed on the bilateral anterior, middle, and posterior parts of the cisternal segment of the trigeminal nerve, the bilateral root entry zone (REZ), bilateral nuclear zone, and the center of pontocerebellar tracts, respectively. Voxel-based morphometry (VBM) analysis was conducted with T2-SPACE images, and gray matter volumes (GMV) were measured from brain pain-matrix regions. Results: The results demonstrated that the VAS scores, the axial diffusivity of the middle part of the affected cisternal trigeminal nerve, the fractional anisotropy of the bilateral nuclear zones, and the mean diffusivity of the center of pontocerebellar tract significantly changed over time before and after treatment. The changes of GMV in the pain-matrix regions exhibited similar trends to the VAS before and after treatment. Conclusion: We conclude that magnetic resonance imaging with RESOLVE DTI and VBM with T2-SPACE images were helpful in the understanding of the pathophysiological mechanisms in patients with TN before and after treatment.

A computerized scheme was developed for automated identification of erect posteroanterior (PA) and supine anteroposterior (AP) chest radiographs. The method was based on three features, the tilt angle of the scapula superior border, the tilt angle of the clavicle and the extent of radiolucence in lung fields, to identify the view of a chest radiograph. The three indices A(scapula), A(clavicle) and C(lung) were determined from a chest image for the three features. Linear discriminant analysis was used to classify PA and AP chest images based on the three indices. The performance of the method was evaluated by receiver operating characteristic analysis. The proposed method was evaluated using a database of 600 PA and 600 AP chest radiographs. The discriminant performances Az of A(scapula), A(clavicle) and C(lung) were 0.878 ± 0.010, 0.683 ± 0.015 and 0.962 ± 0.006, respectively. The combination of the three indices obtained an Az value of 0.979 ± 0.004. The results indicate that the combination of the three indices could yield high discriminant performance. The proposed method could provide radiologists with information about the view of chest radiographs for interpretation or could be used as a preprocessing step for analyzing chest images.