Diffusion Tensor and Kurtosis MRI-Based Radiomics Analysis of Kidney Injury in Type 2 Diabetes.

BACKGROUND: Diffusion tensor imaging (DTI) and diffusion kurtosis imaging (DKI) can provide quantitative parameters that show promise for evaluation of diabetic kidney disease (DKD). The combination of radiomics with DTI and DKI may hold potential clinical value in detecting DKD. PURPOSE: To investigate radiomics models of DKI and DTI for predicting DKD in type 2 diabetes mellitus (T2DM) and evaluate their performance in automated renal parenchyma segmentation. STUDY TYPE: Prospective. POPULATION: One hundred and sixty-three T2DM patients (87 DKD; 63 females; 27-80 years), randomly divided into training cohort (N = 114) and validation cohort (N = 49). FIELD STRENGTH/SEQUENCE: 1.5-T, diffusion spectrum imaging (DSI) with 9 different b-values. ASSESSMENT: The images of DSI were processed to generate DKI and DTI parameter maps, including fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity (RD). The Swin UNETR model was trained with 5-fold cross-validation using 100 samples for renal parenchyma segmentation. Subsequently, radiomics features were automatically extracted from each parameter map. The performance of the radiomics models on the validation cohort was evaluated by utilizing the receiver operating characteristic (ROC) curve. STATISTICAL TESTS: Mann-Whitney U test, Chi-squared test, Pearson correlation coefficient, least absolute shrinkage and selection operator (LASSO), dice similarity coefficient (DSC), decision curve analysis (DCA), area under the curve (AUC), and DeLong's test. The threshold for statistical significance was set at P < 0.05. RESULTS: The DKI_MD achieved the best segmentation performance (DSC, 0.925 ± 0.011). A combined radiomics model (DTI_FA, DTI_MD, DKI_FA, DKI_MD, and DKI_RD) showed the best performance (AUC, 0.918; 95% confidence interval [CI]: 0.820-0.991). When the threshold probability was greater than 20%, the combined model provided the greatest net benefit. Among the single parameter maps, the DTI_FA exhibited superior diagnostic performance (AUC, 887; 95% CI: 0.779-0.972). DATA CONCLUSION: The radiomics signature constructed based on DKI and DTI may be used as an accurate and non-invasive tool to identify T2DM and DKD. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY: Stage 2.

HmsU-Net: A hybrid multi-scale U-net based on a CNN and transformer for medical image segmentation.

Accurate medical image segmentation is of great significance for subsequent diagnosis and analysis. The acquisition of multi-scale information plays an important role in segmenting regions of interest of different sizes. With the emergence of Transformers, numerous networks adopted hybrid structures incorporating Transformers and CNNs to learn multi-scale information. However, the majority of research has focused on the design and composition of CNN and Transformer structures, neglecting the inconsistencies in feature learning between Transformer and CNN. This oversight has resulted in the hybrid network's performance not being fully realized. In this work, we proposed a novel hybrid multi-scale segmentation network named HmsU-Net, which effectively fused multi-scale features. Specifically, HmsU-Net employed a parallel design incorporating both CNN and Transformer architectures. To address the inconsistency in feature learning between CNN and Transformer within the same stage, we proposed the multi-scale feature fusion module. For feature fusion across different stages, we introduced the cross-attention module. Comprehensive experiments conducted on various datasets demonstrate that our approach surpasses current state-of-the-art methods.

SaB-Net: Self-attention backward network for gastric tumor segmentation in CT images.

Gastric cancer is a significant contributor to cancer-related fatalities globally. The automated segmentation of gastric tumors has the potential to analyze the medical condition of patients and enhance the likelihood of surgical treatment success. However, the development of an automatic solution is challenged by the heterogeneous intensity distribution of gastric tumors in computed tomography (CT) images, the low-intensity contrast between organs, and the high variability in the stomach shapes and gastric tumors in different patients. To address these challenges, we propose a self-attention backward network (SaB-Net) for gastric tumor segmentation (GTS) in CT images by introducing a self-attention backward layer (SaB-Layer) to feed the self-attention information learned at the deep layer back to the shallow layers. The SaB-Layer efficiently extracts tumor information from CT images and integrates the information into the network, thereby enhancing the network's tumor segmentation ability. We employed datasets from two centers, one for model training and testing and the other for external validation. The model achieved dice scores of 0.8456 on the test set and 0.8068 on the external verification set. Moreover, we validated the model's transfer learning ability on a publicly available liver cancer dataset, achieving results comparable to state-of-the-art liver cancer segmentation models recently developed. SaB-Net has strong potential for assisting in the clinical diagnosis of and therapy for gastric cancer. Our implementation is available at https://github.com/TyrionJ/SaB-Net.

TGMIL: A hybrid multi-instance learning model based on the Transformer and the Graph Attention Network for whole-slide images classification of renal cell carcinoma.

BACKGROUND AND OBJECTIVES: The pathological diagnosis of renal cell carcinoma is crucial for treatment. Currently, the multi-instance learning method is commonly used for whole-slide image classification of renal cell carcinoma, which is mainly based on the assumption of independent identical distribution. But this is inconsistent with the need to consider the correlation between different instances in the diagnosis process. Furthermore, the problem of high resource consumption of pathology images is still urgent to be solved. Therefore, we propose a new multi-instance learning method to solve this problem. METHODS: In this study, we proposed a hybrid multi-instance learning model based on the Transformer and the Graph Attention Network, called TGMIL, to achieve whole-slide image of renal cell carcinoma classification without pixel-level annotation or region of interest extraction. Our approach is divided into three steps. First, we designed a feature pyramid with the multiple low magnifications of whole-slide image named MMFP. It makes the model incorporates richer information, and reduces memory consumption as well as training time compared to the highest magnification. Second, TGMIL amalgamates the Transformer and the Graph Attention's capabilities, adeptly addressing the loss of instance contextual and spatial. Within the Graph Attention network stream, an easy and efficient approach employing max pooling and mean pooling yields the graph adjacency matrix, devoid of extra memory consumption. Finally, the outputs of two streams of TGMIL are aggregated to achieve the classification of renal cell carcinoma. RESULTS: On the TCGA-RCC validation set, a public dataset for renal cell carcinoma, the area under a receiver operating characteristic (ROC) curve (AUC) and accuracy of TGMIL were 0.98±0.0015,0.9191±0.0062, respectively. It showcased remarkable proficiency on the private validation set of renal cell carcinoma pathology images, attaining AUC of 0.9386±0.0162 and ACC of 0.9197±0.0124. Furthermore, on the public breast cancer whole-slide image test dataset, CAMELYON 16, our model showed good classification performance with an accuracy of 0.8792. CONCLUSIONS: TGMIL models the diagnostic process of pathologists and shows good classification performance on multiple datasets. Concurrently, the MMFP module efficiently diminishes resource requirements, offering a novel angle for exploring computational pathology images.

Combining radiomics and deep learning features of intra-tumoral and peri-tumoral regions for the classification of breast cancer lung metastasis and primary lung cancer with low-dose CT.

PURPOSE: To investigate the performance of deep learning and radiomics features of intra-tumoral region (ITR) and peri-tumoral region (PTR) in the diagnosing of breast cancer lung metastasis (BCLM) and primary lung cancer (PLC) with low-dose CT (LDCT). METHODS: We retrospectively collected the LDCT images of 100 breast cancer patients with lung lesions, comprising 60 cases of BCLM and 40 cases of PLC. We proposed a fusion model that combined deep learning features extracted from ResNet18-based multi-input residual convolution network with traditional radiomics features. Specifically, the fusion model adopted a multi-region strategy, incorporating the aforementioned features from both the ITR and PTR. Then, we randomly divided the dataset into training and validation sets using fivefold cross-validation approach. Comprehensive comparative experiments were performed between the proposed fusion model and other eight models, including the intra-tumoral deep learning model, the intra-tumoral radiomics model, the intra-tumoral deep-learning radiomics model, the peri-tumoral deep learning model, the peri-tumoral radiomics model, the peri-tumoral deep-learning radiomics model, the multi-region radiomics model, and the multi-region deep-learning model. RESULTS: The fusion model developed using deep-learning radiomics feature sets extracted from the ITR and PTR had the best classification performance, with the area under the curve of 0.913 (95% CI 0.840-0.960). This was significantly higher than that of the single region's radiomics model or deep learning model. CONCLUSIONS: The combination of radiomics and deep learning features was effective in discriminating BCLM and PLC. Additionally, the analysis of the PTR can mine more comprehensive tumor information.

msQSM: Morphology-based self-supervised deep learning for quantitative susceptibility mapping.

Quantitative susceptibility mapping (QSM) has been applied to the measurement of iron deposition and the auxiliary diagnosis of neurodegenerative disease. There still exists a dipole inversion problem in QSM reconstruction. Recently, deep learning approaches have been proposed to resolve this problem. However, most of these approaches are supervised methods that need pairs of the input phase and ground-truth. It remains a challenge to train a model for all resolutions without using the ground-truth and only using one resolution data. To address this, we proposed a self-supervised QSM deep learning method based on morphology. It consists of a morphological QSM builder to decouple the dependency of the QSM on acquisition resolution, and a morphological loss to reduce artifacts effectively and save training time efficiently. The proposed method can reconstruct arbitrary resolution QSM on both human data and animal data, regardless of whether the resolution is higher or lower than that of the training set. Our method outperforms the previous best unsupervised method with a 3.6% higher peak signal-to-noise ratio, 16.2% lower normalized root mean square error, and 22.1% lower high-frequency error norm. The morphological loss reduces training time by 22.1% with respect to the cycle gradient loss used in the previous unsupervised methods. Experimental results show that the proposed method accurately measures QSM with arbitrary resolutions, and achieves state-of-the-art results among unsupervised deep learning methods. Research on applications in neurodegenerative diseases found that our method is robust enough to measure significant increase in striatal magnetic susceptibility in patients during Alzheimer's disease progression, as well as significant increase in substantia nigra susceptibility in Parkinson's disease patients, and can be used as an auxiliary differential diagnosis tool for Alzheimer's disease and Parkinson's disease.

Spatiotemporal expression patterns of genes coding for plasmalemmal chloride transporters and channels in neurological diseases.

Neuronal voltage changes which are dependent on chloride transporters and channels are involved in forming neural functions during early development and maintaining their stability until adulthood. The intracellular chloride concentration maintains a steady state, which is delicately regulated by various genes coding for chloride transporters and channels (GClTC) on the plasmalemma; however, the synergistic effect of these genes in central nervous system disorders remains unclear. In this study, we first defined 10 gene clusters with similar temporal expression patterns, and identified 41 GClTC related to brain developmental process. Then, we found 4 clusters containing 22 GClTC were enriched for the neuronal functions. The GClTC from different clusters presented distinct cell type preferences and anatomical heterogeneity. We also observed strong correlations between clustered genes and diseases, most of which were nervous system disorders. Finally, we found that one of the most well-known GClTC, SLC12A2, had a more profound effect on glial cell-related diseases than on neuron-related diseases, which was in accordance with our observation that SLC12A2 was mainly expressed in oligodendrocytes during brain development. Our findings provide a more comprehensive understanding of the temporal and spatial expression characteristics of GClTC, which can help us understand the complex roles of GClTC in the development of the healthy human brain and the etiology of brain disorders.

Multimodal cross enhanced fusion network for diagnosis of Alzheimer's disease and subjective memory complaints.

Deep learning methods using multimodal imagings have been proposed for the diagnosis of Alzheimer's disease (AD) and its early stages (SMC, subjective memory complaints), which may help to slow the progression of the disease through early intervention. However, current fusion methods for multimodal imagings are generally coarse and may lead to suboptimal results through the use of shared extractors or simple downscaling stitching. Another issue with diagnosing brain diseases is that they often affect multiple areas of the brain, making it important to consider potential connections throughout the brain. However, traditional convolutional neural networks (CNNs) may struggle with this issue due to their limited local receptive fields. To address this, many researchers have turned to transformer networks, which can provide global information about the brain but can be computationally intensive and perform poorly on small datasets. In this work, we propose a novel lightweight network called MENet that adaptively recalibrates the multiscale long-range receptive field to localize discriminative brain regions in a computationally efficient manner. Based on this, the network extracts the intensity and location responses between structural magnetic resonance imagings (sMRI) and 18-Fluoro-Deoxy-Glucose Positron Emission computed Tomography (FDG-PET) as an enhancement fusion for AD and SMC diagnosis. Our method is evaluated on the publicly available ADNI datasets and achieves 97.67% accuracy in AD diagnosis tasks and 81.63% accuracy in SMC diagnosis tasks using sMRI and FDG-PET. These results achieve state-of-the-art (SOTA) performance in both tasks. To the best of our knowledge, this is one of the first deep learning research methods for SMC diagnosis with FDG-PET.

Influence of Dehydroxymethylepoxyquinomicin on Radiosensitivity of Thyroid Carcinoma TPC-1 Cells.

Objective: To investigate the influence of dehydroxymethylepoxyquinomicin (DHMEQ), an NF-κB inhibitor, on radiosensitivity of thyroid carcinoma (TC) TPC-1 cells. Methods: The isolation of CDl33 positive cells (CD133+ TPC-1) and negative cells (CD133- TPC-1) from TPC-1 cells used immunomagnetic bead sorting. After verification of the toxicity of DHMEQ to cells by MTT and cell cloning assays, the cells were divided into four groups, of which three groups were intervened by DHMEQ, 131I radiation, and DHMEQ +131I radiation, respectively, while the fourth group was used as a control without treatment. Alterations in cell growth, apoptosis, and cell cycle were observed. Results: DHMEQ had certain toxic effects on TPC-1 cells, with an IC50 of 38.57 µg/mL (P < 0.05). DHMEQ inhibited CD133+ and CD133- TPC-1 proliferation and their clonogenesis after irradiation. DHMEQ + radiation contributed to a growth inhibition rate and an apoptosis rate higher than either or them alone (P < 0.05), with a more significant effect on CD133- TPC-1 than CD133+ TPC-1 under the same treatment conditions (P < 0.05). Conclusion: DHEMQ can increase the radiosensitivity of TC cells to 131I, inhibit tumor cell growth, and promote apoptosis. However, its effect is less significant on CD133+ TPC-1 compared with CD133- TPC-1, which may be related to the stem cell-like properties of CD133+ cells. In the future, the application of DHMEQ in TC 131I radiotherapy will effectively improve the clinical effect of patients.

GEMIN6 Overexpression Correlates with the Low Immune Cell Infiltration and Poor Prognosis in Lung Adenocarcinoma.

Background: Gem nuclear organelle-associated protein 6 (GEMIN6) is a component of the GEMINS protein family involved in the survival of motor neuron (SMN) complex. SMN interfered with snRNP assembly and mRNA processing resulting in tumorigenesis. We performed this study to explore the association between GEMIN6 and lung adenocarcinoma (LUAD). Methods: We used The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases to collect transcriptomic expression data of LUAD patients and analyze the difference in GEMIN6 expression between normal and tumor tissues of LUAD. qRT-PCR analysis was also performed to detect the expression of GEMIN6 in normal and LUAD cells. The expression of GEMIN6 on the LUAD patient survival outcome was estimated by the Kaplan-Meier curves and Cox analyses. In addition, the Metascape online tool and single-sample GSEA were employed to find out the underlying biological mechanisms of GEMIN6. Furthermore, the correlations of GEMIN6 expression with immune cell infiltration in LUAD were analyzed by ssGSEA and Spearman correlation analysis. Results: Compared with the normal tissues and cells, the expression of GEMIN6 was significantly higher in LUAD tissues and cells; the high expression GEMIN6 was also found in the advanced pathologic stage and advanced N and T stages of LUAD. GEMIN6 high expression was significantly associated with inferior overall survival. The heat map revealed the top 20 coexpressed genes with GEMIN6, including SF3B6, CPSF3, and PSMB3. Functional enrichment analysis demonstrated that enrichment genes are associated with the cell cycle, mRNA processing, and energy metabolism. Additionally, GEMIN6 was negatively related to the immune cell infiltration in LUAD. Conclusions: This study demonstrated that GEMIN6 was involved in the tumorigenesis and progression of LUAD. GEMIN6 could be an important molecular marker of poor prognosis and a therapeutic target of LUAD.

Pretreatment DCE-MRI-Based Deep Learning Outperforms Radiomics Analysis in Predicting Pathologic Complete Response to Neoadjuvant Chemotherapy in Breast Cancer.

Purpose: To compare the performances of deep learning (DL) to radiomics analysis (RA) in predicting pathological complete response (pCR) to neoadjuvant chemotherapy (NAC) based on pretreatment dynamic contrast-enhanced MRI (DCE-MRI) in breast cancer. Materials and Methods: This retrospective study included 356 breast cancer patients who underwent DCE-MRI before NAC and underwent surgery after NAC. Image features and kinetic parameters of tumors were derived from DCE-MRI. Molecular information was assessed based on immunohistochemistry results. The image-based RA and DL models were constructed by adding kinetic parameters or molecular information to image-only linear discriminant analysis (LDA) and convolutional neural network (CNN) models. The predictive performances of developed models were assessed by receiver operating characteristic (ROC) curve analysis and compared with the DeLong method. Results: The overall pCR rate was 23.3% (83/356). The area under the ROC (AUROC) of the image-kinetic-molecular RA model was 0.781 [95% confidence interval (CI): 0.735, 0.828], which was higher than that of the image-kinetic RA model (0.629, 95% CI: 0.595, 0.663; P < 0.001) and comparable to that of the image-molecular RA model (0.755, 95% CI: 0.708, 0.802; P = 0.133). The AUROC of the image-kinetic-molecular DL model was 0.83 (95% CI: 0.816, 0.847), which was higher than that of the image-kinetic and image-molecular DL models (0.707, 95% CI: 0.654, 0.761; 0.79, 95% CI: 0.768, 0.812; P < 0.001) and higher than that of the image-kinetic-molecular RA model (0.778, 95% CI: 0.735, 0.828; P < 0.001). Conclusions: The pretreatment DCE-MRI-based DL model is superior to the RA model in predicting pCR to NAC in breast cancer patients. The image-kinetic-molecular DL model has the best prediction performance.

BMNet: A New Region-Based Metric Learning Method for Early Alzheimer's Disease Identification With FDG-PET Images.

18F-fluorodeoxyglucose (FDG)-positron emission tomography (PET) reveals altered brain metabolism in individuals with mild cognitive impairment (MCI) and Alzheimer's disease (AD). Some biomarkers derived from FDG-PET by computer-aided-diagnosis (CAD) technologies have been proved that they can accurately diagnosis normal control (NC), MCI, and AD. However, existing FDG-PET-based researches are still insufficient for the identification of early MCI (EMCI) and late MCI (LMCI). Compared with methods based other modalities, current methods with FDG-PET are also inadequate in using the inter-region-based features for the diagnosis of early AD. Moreover, considering the variability in different individuals, some hard samples which are very similar with both two classes limit the classification performance. To tackle these problems, in this paper, we propose a novel bilinear pooling and metric learning network (BMNet), which can extract the inter-region representation features and distinguish hard samples by constructing the embedding space. To validate the proposed method, we collect 898 FDG-PET images from Alzheimer's disease neuroimaging initiative (ADNI) including 263 normal control (NC) patients, 290 EMCI patients, 147 LMCI patients, and 198 AD patients. Following the common preprocessing steps, 90 features are extracted from each FDG-PET image according to the automatic anatomical landmark (AAL) template and then sent into the proposed network. Extensive fivefold cross-validation experiments are performed for multiple two-class classifications. Experiments show that most metrics are improved after adding the bilinear pooling module and metric losses to the Baseline model respectively. Specifically, in the classification task between EMCI and LMCI, the specificity improves 6.38% after adding the triple metric loss, and the negative predictive value (NPV) improves 3.45% after using the bilinear pooling module. In addition, the accuracy of classification between EMCI and LMCI achieves 79.64% using imbalanced FDG-PET images, which illustrates that the proposed method yields a state-of-the-art result of the classification accuracy between EMCI and LMCI based on PET images.

FMRNet: A fused network of multiple tumoral regions for breast tumor classification with ultrasound images.

PURPOSE: Recent studies have illustrated that the peritumoral regions of medical images have value for clinical diagnosis. However, the existing approaches using peritumoral regions mainly focus on the diagnostic capability of the single region and ignore the advantages of effectively fusing the intratumoral and peritumoral regions. In addition, these methods need accurate segmentation masks in the testing stage, which are tedious and inconvenient in clinical applications. To address these issues, we construct a deep convolutional neural network that can adaptively fuse the information of multiple tumoral-regions (FMRNet) for breast tumor classification using ultrasound (US) images without segmentation masks in the testing stage. METHODS: To sufficiently excavate the potential relationship, we design a fused network and two independent modules to extract and fuse features of multiple regions simultaneously. First, we introduce two enhanced combined-tumoral (EC) region modules, aiming to enhance the combined-tumoral features gradually. Then, we further design a three-branch module for extracting and fusing the features of intratumoral, peritumoral, and combined-tumoral regions, denoted as the intratumoral, peritumoral, and combined-tumoral module. Especially, we design a novel fusion module by introducing a channel attention module to adaptively fuse the features of three regions. The model is evaluated on two public datasets including UDIAT and BUSI with breast tumor ultrasound images. Two independent groups of experiments are performed on two respective datasets using the fivefold stratified cross-validation strategy. Finally, we conduct ablation experiments on two datasets, in which BUSI is used as the training set and UDIAT is used as the testing set. RESULTS: We conduct detailed ablation experiments about the proposed two modules and comparative experiments with other existing representative methods. The experimental results show that the proposed method yields state-of-the-art performance on both two datasets. Especially, in the UDIAT dataset, the proposed FMRNet achieves a high accuracy of 0.945 and a specificity of 0.945, respectively. Moreover, the precision (PRE = 0.909) even dramatically improves by 21.6% on the BUSI dataset compared with the existing method of the best result. CONCLUSION: The proposed FMRNet shows good performance in breast tumor classification with US images, and proves its capability of exploiting and fusing the information of multiple tumoral-regions. Furthermore, the FMRNet has potential value in classifying other types of cancers using multiple tumoral-regions of other kinds of medical images.

Design of a rapid diagnostic model for bladder compliance based on real-time intravesical pressure monitoring system.

OBJECTIVE: The diagnosis of bladder dysfunction for children depends on the confirmation of abnormal bladder shape and bladder compliance. The existing gold standard needs to conduct voiding cystourethrogram (VCUG) examination and urodynamic studies (UDS) examination on patients separately. To reduce the time and injury of children's inspection, we propose a novel method to judge the bladder compliance by measuring the intravesical pressure during the VCUG examination without extra UDS. METHODS: Our method consisted of four steps. We firstly developed a single-tube device that can measure, display, store, and transmit real-time pressure data. Secondly, we conducted clinical trials with the equipment on a cohort of 52 patients (including 32 negative and 20 positive cases). Thirdly, we preprocessed the data to eliminate noise and extracted features, then we used the least absolute shrinkage and selection operator (LASSO) to screen out important features. Finally, several machine learning methods were applied to classify and predict the bladder compliance level, including support vector machine (SVM), Random Forest, XGBoost, perceptron, logistic regression, and Naive Bayes, and the classification performance was evaluated. RESULTS: 73 features were extracted, including first-order and second-order time-domain features, wavelet features, and frequency domain features. 15 key features were selected and the model showed promising classification performance. The highest AUC value was 0.873 by the SVM algorithm, and the corresponding accuracy was 84%. CONCLUSION: We designed a system to quickly obtain the intravesical pressure during the VCUG test, and our classification model is competitive in judging patients' bladder compliance. SIGNIFICANCE: This could facilitate rapid auxiliary diagnosis of bladder disease based on real-time data. The promising result of classification is expected to provide doctors with a reliable basis in the auxiliary diagnosis of some bladder diseases prior to UDS.

Classification of microcalcification clusters in digital breast tomosynthesis using ensemble convolutional neural network.

BACKGROUND: The classification of benign and malignant microcalcification clusters (MCs) is an important task for computer-aided diagnosis (CAD) of digital breast tomosynthesis (DBT) images. Influenced by imaging method, DBT has the characteristic of anisotropic resolution, in which the resolution of intra-slice and inter-slice is quite different. In addition, the sharpness of MCs in different slices of DBT is quite different, among which the clearest slice is called focus slice. These characteristics limit the performance of CAD algorithms based on standard 3D convolution neural network (CNN). METHODS: To make full use of the characteristics of the DBT, we proposed a new ensemble CNN, which consists of the 2D ResNet34 and the anisotropic 3D ResNet to extract the 2D focus slice features and 3D contextual features of MCs, respectively. Moreover, the anisotropic 3D convolution is used to build 3D ResNet to avoid the influence of DBT anisotropy. RESULTS: The proposed method was evaluated on 495 MCs in DBT images of 275 patients, which are collected from our collaborative hospital. The area under the curve (AUC) of receiver operating characteristic (ROC) and accuracy of classifying benign and malignant MCs using decision-level ensemble strategy were 0.8837 and 82.00%, which were significantly higher than the experimental results of 2D ResNet34 (AUC: 0.8264, ACC: 76.00%) and anisotropic 3D ResNet (AUC: 0.8455, ACC: 76.00%). Compared with the results of 3D features classification in the radiomics, the AUC of the deep learning method with decision-level ensemble strategy was improved by 0.0435, and the F1 score was improved from 79.37 to 85.71%. More importantly, the sensitivity increased from 78.13 to 84.38%, and the specificity increased from 66.67 to 77.78%, which effectively reduced the false positives of diagnosis CONCLUSION: The results fully prove that the ensemble CNN can effectively integrate 2D features and 3D features, improve the classification performance of benign and malignant MCs in DBT, and reduce the false positives.

3D Context-Aware Convolutional Neural Network for False Positive Reduction in Clustered Microcalcifications Detection.

False positives (FPs) reduction is indispensable for clustered microcalcifications (MCs) detection in digital breast tomosynthesis (DBT), since there might be excessive false candidates in the detection stage. Considering that DBT volume has an anisotropic resolution, we proposed a novel 3D context-aware convolutional neural network (CNN) to reduce FPs, which consists of a 2D intra-slices feature extraction branch and a 3D inter-slice features fusion branch. In particular, 3D anisotropic convolutions were designed to learn representations from DBT volumes and inter-slice information fusion is only performed on the feature map level, which could avoid the influence of anisotropic resolution of DBT volume. The proposed method was evaluated on a large-scale Chinese women population of 877 cases with 1754 DBT volumes and compared with 8 related methods. Experimental results show that the proposed network achieved the best performance with an accuracy of 92.68% for FPs reduction with an AUC of 97.65%, and the FPs are 0.0512 per DBT volume at a sensitivity of 90%. This also proved that making full use of 3D contextual information of DBT volume can improve the performance of the classification algorithm.

A radiomics method to classify microcalcification clusters in digital breast tomosynthesis.

PURPOSE: Digital breast tomosynthesis (DBT) is becoming increasingly used in clinical practice. In DBT, the microcalcification clusters may span across multiple slices, which makes it difficult for radiologists to directly assess these distributed clusters for diagnosis. We investigated a radiomics method to classify microcalcification clusters in DBT based on a semiautomatic segmentation process. METHODS: We performed a retrospective study on a cohort of 275 patients (including 79 benign and 196 malignant cases) with a total of 550 DBT volumes. Our method consisted of three steps. The initial step was to semiautomatically segment the microcalcification clusters. Then, radiomics features were extracted from the initially segmented microcalcification clusters. Finally, the benign and malignant microcalcification clusters were differentiated by the random forest (RF) classifier using selected subset features. The radiomics models were evaluated both on view-based and case-based modes with features selected from different domains. The receiver operating characteristic (ROC) curve and area under the ROC curve (AUC) were used to evaluate the classification performance. RESULTS: Twenty-six key features were selected from a total of 170 radiomics features and these features show promising classification performance. The highest AUC was 0.834 for view-based mode and 0.868 for case-based mode when using features selected from the 3D-domain. The 2D-domain radiomics features showed a statistically similar performance to the 3D features (P > 0.05). CONCLUSION: Radiomics models can provide encouraging performance in classification between malignant and benign microcalcification clusters which are semiautomatically segmented in DBT.

Radiomics Based on Multimodal MRI for the Differential Diagnosis of Benign and Malignant Breast Lesions.

BACKGROUND: MRI-based radiomics has been used to diagnose breast lesions; however, little research combining quantitative pharmacokinetic parameters of dynamic contrast-enhanced MRI (DCE-MRI) and diffusion kurtosis imaging (DKI) exists. PURPOSE: To develop and validate a multimodal MRI-based radiomics model for the differential diagnosis of benign and malignant breast lesions and analyze the discriminative abilities of different MR sequences. STUDY TYPE: Retrospective. POPULATION: In all, 207 female patients with 207 histopathology-confirmed breast lesions (95 benign and 112 malignant) were included in the study. Then 159 patients were assigned to the training group, and 48 patients comprised the validation group. FIELD STRENGTH/SEQUENCE: T2 -weighted (T2 W), T1 -weighted (T1 W), diffusion-weighted MR imaging (b-values = 0, 500, 800, and 2000 seconds/mm2 ) and quantitative DCE-MRI were performed on a 3.0T MR scanner. ASSESSMENT: Radiomics features were extracted from T2 WI, T1 WI, DKI, apparent diffusion coefficient (ADC) maps, and DCE pharmacokinetic parameter maps in the training set. Models based on each sequence or combinations of sequences were built using a support vector machine (SVM) classifier and used to differentiate benign and malignant breast lesions in the validation set. STATISTICAL TESTS: Optimal feature selection was performed by Spearman's rank correlation coefficients and the least absolute shrinkage and selection operator algorithm (LASSO). Receiver operating characteristic (ROC) curves were used to assess the diagnostic performance of the radiomics models in the validation set. RESULTS: The area under the ROC curve (AUC) of the optimal radiomics model, including T2 WI, DKI, and quantitative DCE-MRI parameter maps was 0.921, with an accuracy of 0.833. The AUCs of the models based on T1 WI, T2 WI, ADC map, DKI, and DCE pharmacokinetic parameter maps were 0.730, 0.791, 0.770, 0.788, and 0.836, respectively. DATA CONCLUSION: The model based on radiomics features from T2 WI, DKI, and quantitative DCE pharmacokinetic parameter maps has a high discriminatory ability for benign and malignant breast lesions. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY STAGE: 2 J. Magn. Reson. Imaging 2020;52:596-607.

Impact of total parenteral nutrition standardization led by pharmacist on quality in postoperative patients with colorectal cancer.

BACKGROUND/OBJECTIVES: Abdominal surgery significantly affects the structure and function of the gastrointestinal system of patients, total parenteral nutrition (TPN) is an important nutrition support method for postoperative patients. However, in the process of TPN practice, the excessive fat emulsion and compound amino-acid prescriptions ratio are often prescribed by doctors. To address the problem, we developed the computerized TPN prescription management system to promote the personalized provision of TPN. The purpose of this study is to evaluate the intervention effects of the computerized TPN prescription management system, which is designed by pharmacists in the Surgical Department of Abdominal Oncology at Zhejiang Cancer Hospital in July 2015. SUBJECTS/METHODS: The computerized TPN prescription management system applied in Surgical Department of Abdominal Oncology on 1 July 2015. The computerized TPN prescription management system was evaluated by comparing the patients who were treated 3 months after the application of the system with the control subjects who were treated 3 months prior to the application of TPN prescription management system in Surgical Department of Abdominal Oncology. RESULTS: In total, 218 TPN prescription-treated patients with colorectal cancer received surgery treatment were analyzed, including 121 subjects who received the treatment 3 months prior to application of TPN prescription system (IPN period) and 97 subjects who received the treatment after 3 months of the system application (SPN period). The rates of optimized TPN prescriptions are 47.1% and 88.7% prior to and after application of TPN prescription review system, respectively (p < 0.001). In detail, prior to application of TPN prescription review system, abnormal glucose-lipid ratio and nitrogen-calorie ratio are the most common problems, which accounted for 74.3 and 97.9%, respectively (p < 0.01). Whereas the proportion of the insufficient dosage of amino acids is 62 and 96.9%, respectively (p < 0.01). Other problems are insufficient dosage of insulin and excessive fat soluble vitamin supplement. After application of TPN prescription review system, as the glucose-lipid ratio and nitrogen-calorie ratio are set up in fixed range according to the nutrition treatment guidelines, only a small amount of TPN prescriptions have the problem of insufficient dosage of compound amino acid. Furthermore, before and after the application of TPN management software, the gender, age, performance status (PS) score and BMI index of the two groups of colorectal cancer patients were not statistically different (p > 0.05). There were significant differences in albumin and prealbumin between the two groups after operation (p < 0.05), and there was a significant difference in total protein (p < 0.001). There were significant differences in alanine aminotransferase and indirect bilirubin between liver and kidney function (p < 0.01), and there were significant differences in aspartate aminotransferase and total bilirubin (p < 0.05). Other total cholesterol, L-γ-glutamyl transferase, direct bilirubin and creatinine were not statistically different (p > 0.05). Blood routine (WBC, Hb and lymphocyte), length of stay and recurrence rate were not statistically different (p > 0.05). CONCLUSIONS: The application of TPN management software not only standardized the doctor's TPN medical advice, but also improved the qualified rate of TPN doctor's advice, thus ensuring the safety of the patient's medication. It also had a positive effect on postoperative recovery of colorectal cancer patients, and ensured the efficacy of the treatment of patients. In addition, it reduced the workload of the pharmacist's audit prescription and improved the efficiency of the audit prescription, and further emphasized the role and value of pharmacists.