Diagnosis for autism spectrum disorder children using T1-based gray matter and arterial spin labeling-based cerebral blood flow network metrics.

Introduction: Autism Spectrum Disorder (ASD) is a complex neurodevelopmental condition characterized by impairments in motor skills, communication, emotional expression, and social interaction. Accurate diagnosis of ASD remains challenging due to the reliance on subjective behavioral observations and assessment scales, lacking objective diagnostic indicators. Methods: In this study, we introduced a novel approach for diagnosing ASD, leveraging T1-based gray matter and ASL-based cerebral blood flow network metrics. Thirty preschool-aged patients with ASD and twenty-two typically developing (TD) individuals were enrolled. Brain network features, including gray matter and cerebral blood flow metrics, were extracted from both T1-weighted magnetic resonance imaging (MRI) and ASL images. Feature selection was performed using statistical t-tests and Minimum Redundancy Maximum Relevance (mRMR). A machine learning model based on random vector functional link network was constructed for diagnosis. Results: The proposed approach demonstrated a classification accuracy of 84.91% in distinguishing ASD from TD. Key discriminating network features were identified in the inferior frontal gyrus and superior occipital gyrus, regions critical for social and executive functions in ASD patients. Discussion: Our study presents an objective and effective approach to the clinical diagnosis of ASD, overcoming the limitations of subjective behavioral observations. The identified brain network features provide insights into the neurobiological mechanisms underlying ASD, potentially leading to more targeted interventions.

A Novel Mis-Seg-Focus Loss Function Based on a Two-Stage nnU-Net Framework for Accurate Brain Tissue Segmentation.

Brain tissue segmentation plays a critical role in the diagnosis, treatment, and study of brain diseases. Accurately identifying these boundaries is essential for improving segmentation accuracy. However, distinguishing boundaries between different brain tissues can be challenging, as they often overlap. Existing deep learning methods primarily calculate the overall segmentation results without adequately addressing local regions, leading to error propagation and mis-segmentation along boundaries. In this study, we propose a novel mis-segmentation-focused loss function based on a two-stage nnU-Net framework. Our approach aims to enhance the model's ability to handle ambiguous boundaries and overlapping anatomical structures, thereby achieving more accurate brain tissue segmentation results. Specifically, the first stage targets the identification of mis-segmentation regions using a global loss function, while the second stage involves defining a mis-segmentation loss function to adaptively adjust the model, thus improving its capability to handle ambiguous boundaries and overlapping anatomical structures. Experimental evaluations on two datasets demonstrate that our proposed method outperforms existing approaches both quantitatively and qualitatively.

Uncertain prediction of deformable image registration on lung CT using multi-category features and supervised learning.

The assessment of deformable registration uncertainty is an important task for the safety and reliability of registration methods in clinical applications. However, it is typically done by a manual and time-consuming procedure. We propose a novel automatic method to predict registration uncertainty based on multi-category features and supervised learning. Three types of features, including deformation field statistical features, deformation field physiologically realistic features, and image similarity features, are introduced and calculated to train the random forest regressor for local registration uncertain prediction. Deformation field statistical features represent the numerical stability of registration optimization, which are correlated to the uncertainty of deformation fields; deformation field physiologically realistic features represent the biomechanical properties of organ motions, which mathematically reflect the physiological reality of deformation; image similarity features reflect the similarity between the warped image and fixed image. The multi-category features comprehensively reflect the registration uncertainty. The strategy of spatial adaptive random perturbations is also introduced to accurately simulate spatial distribution of registration uncertainty, which makes deformation field statistical features more discriminative to the uncertainty of deformation fields. Experiments were conducted on three publicly available thoracic CT image datasets. Seventeen randomly selected image pairs are used to train the random forest model, and 9 image pairs are used to evaluate the prediction model. The quantitative experiments on lung CT images show that the proposed method outperforms the baseline method for uncertain prediction of classical iterative optimization-based registration and deep learning-based registration with different registration qualities. The proposed method achieves good performance for registration uncertain prediction, which has great potential in improving the accuracy of registration uncertain prediction.

Quantifying cerebral blood flow changes using arterial spin labeling: A comparative study of idiopathic rapid eye movement sleep behavior disorder and Parkinson's disease.

INTRODUCTION: Idiopathic rapid eye movement sleep behavior disorder (iRBD) and Parkinson's disease (PD) have been found to have changes in cerebral perfusion and overlap of some of the lesioned brain areas. However, a consensus regarding the specific location and diagnostic significance of these cerebral blood perfusion alternations remains elusive in both iRBD and PD. The present study evaluated the patterns of cerebral blood flow changes in iRBD and PD. MATERIAL AND METHODS: A total of 59 right-handed subjects were enrolled, including 15 patients with iRBD, 20 patients with PD, and 24 healthy controls (HC). They were randomly divided into groups at a ratio of 4 to 1 for training and testing. A PASL sequence was employed to obtain quantitative cerebral blood flow (CBF) maps. The CBF values were calculated from these acquired maps. In addition, AutoGluon was employed to construct a classifier for CBF features selection and classification. An independent t-test was performed for CBF variations, with age and sex as nuisance variables. The performance of the feature was evaluated using receiver operating characteristic (ROC) curves. A significance level of P < 0.05 was considered significant. CBF in several brain regions, including the left median cingulate and paracingulate gyri and the right middle occipital gyrus (MOG), showed significant differences between PD and HC, demonstrating good classification performance. The combined model that integrates all features achieved even higher performance with an AUC of 0.9380. Additionally, CBF values in multiple brain regions, including the right MOG and the left angular gyrus, displayed significant differences between PD and iRBD. Particularly, CBF values in the left angular gyrus exhibited good performance in classifying PD and iRBD. The combined model achieved improved performance, with an AUC of 0.8533. No significant differences were found in brain regions when comparing CBF values between iRBD and HC subjects. CONCLUSIONS: ASL-based quantitative CBF change features can offer reliable biomarkers to assist in the diagnosis of PD. Regarding the characteristic of CBF in the right MOG, it is anticipated to serve as an imaging biomarker for predicting the progression of iRBD to PD.

Flexible-center Hat Complete Electrode Model for EEG Forward Problem.

OBJECTIVE: This study aims to develop a more realistic electrode model by incorporating the non-uniform distribution of electrode contact conductance (ECC) and the shunting effects, to accurately solve EEG forward problem (FP). METHODS: Firstly, a hat function is introduced to construct a more realistic hat-shaped distribution (HD) for ECC. Secondly, this hat function is modified by applying two parameters - offset ratio and offset direction - to account for the variability in ECC's center and to develop the flexible-center HD (FCHD). Finally, by integrating this FCHD into the complete electrode model (CEM) with the shunting effects, a novel flexible-center hat complete electrode model (FCH-CEM) is proposed and used to solve FP. RESULTS: Simulation experiments using a realistic head model demonstrate the necessity of FCHCEM and its potential to improve the accuracy of the FP solution compared to current models, i.e., the point electrode model (PEM) and CEM. And compared to PEM, it has better performance under coarse mesh conditions (2 mm). Further experiments indicate the significance of considering shunting effects, as ignoring them results in larger errors than coarse mesh when the average contact conductance is large (). CONCLUSION: The proposed FCH-CEM has better accuracy and performance than PEM and complements CEM in finer meshes, making it necessary for coarse meshes. SIGNIFICANCE: This study proposes a novel model that enhances electrode modeling and FP accuracy, and provides new ideas and methods for future research.

Automated detection of MRI-negative temporal lobe epilepsy with ROI-based morphometric features and machine learning.

Objective: Temporal lobe epilepsy (TLE) predominantly originates from the anteromedial basal region of the temporal lobe, and its prognosis is generally favorable following surgical intervention. However, TLE often appears negative in magnetic resonance imaging (MRI), making it difficult to quantitatively diagnose the condition solely based on clinical symptoms. There is a pressing need for a quantitative, automated method for detecting TLE. Methods: This study employed MRI scans and clinical data from 51 retrospective epilepsy cases, dividing them into two groups: 34 patients in TLE group and 17 patients in non-TLE group. The criteria for defining the TLE group were successful surgical removal of the epileptogenic zone in the temporal lobe and a favorable postoperative prognosis. A standard procedure was used for normalization, brain extraction, tissue segmentation, regional brain partitioning, and cortical reconstruction of T1 structural MRI images. Morphometric features such as gray matter volume, cortical thickness, and surface area were extracted from a total of 20 temporal lobe regions in both hemispheres. Support vector machine (SVM), extreme learning machine (ELM), and cmcRVFL+ classifiers were employed for model training and validated using 10-fold cross-validation. Results: The results demonstrated that employing ELM classifiers in conjunction with specific temporal lobe gray matter volume features led to a better identification of TLE. The classification accuracy was 92.79%, with an area under the curve (AUC) value of 0.8019. Conclusion: The method proposed in this study can significantly assist in the preoperative identification of TLE patients. By employing this method, TLE can be included in surgical criteria, which could alleviate patient symptoms and improve prognosis, thereby bearing substantial clinical significance.

Narrow-band loss - a novel loss function focused on target boundary.

Loss functions widely employed in medical image segmentation, e.g., Dice or Generalized Dice, treat each pixel of segmentation target(s) equally. These region-based loss functions are concerned with the overall segmentation accuracy. However, in clinical applications, the focus of attention is often the boundary area of the target organ(s). Existing region-based loss functions lack attention to boundary areas. We designed narrow-band loss, which computes the integration of the predicted probability within the narrow-band around the target boundary. From the aspect of how it's derived, Narrow-band loss belongs to the region-based loss function. The difference from normal region-based loss is that Narrow-band loss calculates based on the degree of coincidence of the region surrounding the organ boundary. The advantage is that narrow-band loss can guide networks to focus on the target's boundary and neighborhood. We also generalize narrow-band loss to multi-target segmentation. We tested narrow-band loss on two datasets of different parts of the human body: the brain dataset with 416 cases, each case with 35 labels, and the abdominal dataset with 50 cases, each case with 12 labels. Narrow-band loss has improved greatly in hd95 metric and dice metric compared with baseline, which is dice loss only.

An artificial intelligence-assisted diagnosis modeling software (AIMS) platform based on medical images and machine learning: a development and validation study.

Background: Supervised machine learning methods [both radiomics and convolutional neural network (CNN)-based deep learning] are usually employed to develop artificial intelligence models with medical images for computer-assisted diagnosis and prognosis of diseases. A classical machine learning-based modeling workflow involves a series of interconnected components and various algorithms, but this makes it challenging, tedious, and labor intensive for radiologists and researchers to build customized models for specific clinical applications if they lack expertise in machine learning methods. Methods: We developed a user-friendly artificial intelligence-assisted diagnosis modeling software (AIMS) platform, which supplies standardized machine learning-based modeling workflows for computer-assisted diagnosis and prognosis systems with medical images. In contrast to other existing software platforms, AIMS contains both radiomics and CNN-based deep learning workflows, making it an all-in-one software platform for machine learning-based medical image analysis. The modular design of AIMS allows users to build machine learning models easily, test models comprehensively, and fairly compare the performance of different models in a specific application. The graphical user interface (GUI) enables users to process large numbers of medical images without programming or script addition. Furthermore, AIMS also provides a flexible image processing toolkit (e.g., semiautomatic segmentation, registration, morphological operations) to rapidly create lesion labels for multiphase analysis, multiregion analysis of an individual tumor (e.g., tumor mass and peritumor), and multimodality analysis. Results: The functionality and efficiency of AIMS were demonstrated in 3 independent experiments in radiation oncology, where multiphase, multiregion, and multimodality analyses were performed, respectively. For clear cell renal cell carcinoma (ccRCC) Fuhrman grading with multiphase analysis (sample size =187), the area under the curve (AUC) value of the AIMS was 0.776; for ccRCC Fuhrman grading with multiregion analysis (sample size =177), the AUC value of the AIMS was 0.848; for prostate cancer Gleason grading with multimodality analysis (sample size =206), the AUC value of the AIMS was 0.980. Conclusions: AIMS provides a user-friendly infrastructure for radiologists and researchers, lowering the barrier to building customized machine learning-based computer-assisted diagnosis models for medical image analysis.

Multi-phase-combined CECT radiomics models for Fuhrman grade prediction of clear cell renal cell carcinoma.

Objective: This study aimed to evaluate the effectiveness of multi-phase-combined contrast-enhanced CT (CECT) radiomics methods for noninvasive Fuhrman grade prediction of clear cell renal cell carcinoma (ccRCC). Methods: A total of 187 patients with four-phase CECT images were retrospectively enrolled and then were categorized into training cohort (n=126) and testing cohort (n=61). All patients were confirmed as ccRCC by histopathological reports. A total of 110 3D classical radiomics features were extracted from each phase of CECT for individual ccRCC lesion, and contrast-enhanced variation features were also calculated as derived radiomics features. These features were concatenated together, and redundant features were removed by Pearson correlation analysis. The discriminative features were selected by minimum redundancy maximum relevance method (mRMR) and then input into a C-support vector classifier to build multi-phase-combined CECT radiomics models. The prediction performance was evaluated by the area under the curve (AUC) of receiver operating characteristic (ROC). Results: The multi-phase-combined CECT radiomics model showed the best prediction performance (AUC=0.777) than the single-phase CECT radiomics model (AUC=0.711) in the testing cohort (p value=0.039). Conclusion: The multi-phase-combined CECT radiomics model is a potential effective way to noninvasively predict Fuhrman grade of ccRCC. The concatenation of first-order features and texture features extracted from corticomedullary phase and nephrographic phase are discriminative feature representations.

macJNet: weakly-supervised multimodal image deformable registration using joint learning framework and multi-sampling cascaded MIND.

Deformable multimodal image registration plays a key role in medical image analysis. It remains a challenge to find accurate dense correspondences between multimodal images due to the significant intensity distortion and the large deformation. macJNet is proposed to align the multimodal medical images, which is a weakly-supervised multimodal image deformable registration method using a joint learning framework and multi-sampling cascaded modality independent neighborhood descriptor (macMIND). The joint learning framework consists of a multimodal image registration network and two segmentation networks. The proposed macMIND is a modality-independent image structure descriptor to provide dense correspondence for registration, which incorporates multi-orientation and multi-scale sampling patterns to build self-similarity context. It greatly enhances the representation ability of cross-modal features in the registration network. The semi-supervised segmentation networks generate anatomical labels to provide semantics correspondence for registration, and the registration network helps to improve the performance of multimodal image segmentation by providing the consistency of anatomical labels. 3D CT-MR liver image dataset with 118 samples is built for evaluation, and comprehensive experiments have been conducted to demonstrate that macJNet achieves superior performance over state-of-the-art multi-modality medical image registration methods.

Unsupervised registration for liver CT-MR images based on the multiscale integrated spatial-weight module and dual similarity guidance.

PURPOSE: Multimodal registration is a key task in medical image analysis. Due to the large differences of multimodal images in intensity scale and texture pattern, it is a great challenge to design distinctive similarity metrics to guide deep learning-based multimodal image registration. Besides, since the limitation of the small receptive field, existing deep learning-based methods are mainly suitable for small deformation, but helpless for large deformation. To address the above issues, we present an unsupervised multimodal image registration method based on the multiscale integrated spatial-weight module and dual similarity guidance. METHODS: In this method, a U-shape network with our multiscale integrated spatial-weight module is embedded into a multi-resolution image registration architecture to achieve end-to-end large deformation registration, where the spatial-weight module can effectively highlight the regions with large deformation and aggregate discriminative features, and the multi-resolution architecture further helps to solve the optimization problem of the network in a coarse-to-fine pattern. Furthermore, we introduce a special loss function based on dual similarity, which represents both global gray-scale similarity and local feature similarity, to optimize the unsupervised multimodal registration network. RESULTS: We verified the effectiveness of the proposed method on liver CT-MR images. Experimental results indicate that the proposed method achieves the optimal DSC value and TRE value of 92.70 ± 1.75(%) and 6.52 ± 2.94(mm), compared with other state-of-the-art registration algorithms. CONCLUSION: The proposed method can accurately estimate the large deformation field by aggregating multiscale features, and achieve higher registration accuracy and fast registration speed. Comparative experiments also demonstrate the effectiveness and generalization ability of the algorithm.

Using radiomics based on multicenter magnetic resonance images to predict isocitrate dehydrogenase mutation status of gliomas.

Background: Isocitrate dehydrogenase (IDH) mutation status is an important biomarker for the treatment strategy selection and prognosis evaluation of glioma. The purpose of this study is to predict the IDH mutation status of gliomas based on multicenter magnetic resonance (MR) images using radiomic models, which were composed from the selected radiomics features and logistic regression (LR), support vector machine (SVM), and LR least absolute shrinkage and selection operator (LASSO) classifiers. Methods: We retrospectively reviewed the medical records of 205 patients with gliomas. We enrolled 78 patients from Shandong Provincial Hospital from January 2018 to December 2019 as testing sets and 127 patients from The Cancer Genome Atlas (TCGA) as training sets. Preoperative MR images were stratified according to their IDH status, and the participants formed a consecutive and random series. Four MR modalities, including T1C, T2, T1 fluid-attenuated inversion recovery (FLAIR), and T2 FLAIR, were used for analysis. Five-fold cross-validation was adopted to train the models, and the models' performances were verified through the testing set. Tumor volumes of interest (VOI) were delineated on the 4 MR modalities. A total of 428 radiomics features were extracted. Two feature selection algorithms, Pearson correlation coefficient (PCC) and recursive feature elimination (RFE), were used to select radiomics features. These features were fed into 3 machine learning classifiers, which were LR, SVM, and LR LASSO, to construct prediction models. The accuracy (ACC), sensitivity (SEN), specificity (SPEC), and area under the curve (AUC) were applied to measure the predictive performance of the radiomics models. Results: The LR (SVM and LR LASSO) classifier predicted IDH mutation status with an average testing set ACC of 80.77% (80.64% and 80.41%), a SEN of 73.68% (84.21% and 89.47%), a SPEC of 87.50% (67.50% and 62.50%), and an AUC of 0.8572 (0.8217 and 0.8164). Conclusions: The radiomics models based on MR modalities demonstrated the potential to be used as tools across different data sets for the noninvasive prediction of the IDH mutation status in glioma.

Integration of simultaneous fMRI and EEG source localization in emotional decision problems.

Simultaneous EEG-fMRI has been a powerful technique to understand the mechanism of the brain in recent years. In this paper, we develop an integrating method by integrating the EEG data into the fMRI data based on the parametric empirical Bayesian (PEB) model to improve the accuracy of the brain source location. The gambling task, a classic paradigm, is used for the emotional decision-making study in this paper. The proposed method was conducted on 21 participants, including 16 men and 5 women. Contrary to the previous method that only localizes the area widely distributed across the ventral striatum and orbitofrontal cortex, the proposed method localizes accurately at the orbital frontal cortex during the process of the brain's emotional decision-making. The activated brain regions extracted by source localization were mainly located in the prefrontal and orbitofrontal lobes; the activation of the temporal pole regions unrelated to reward processing disappeared, and the activation of the somatosensory cortex and motor cortex was significantly reduced. The log evidence shows that the integration of simultaneous fMRI and EEG method based on synchronized data evidence is 22,420, the largest value among the three methods. The integration method always takes on a larger value of log evidence and describes a better performance in analysis associated with source localization. DATA AVAILABILITY: The data used in the current study are available from the corresponding author upon on reasonable request.

A Cross-Space CNN With Customized Characteristics for Motor Imagery EEG Classification.

The classification of motor imagery-electroencephalogram(MI-EEG)based brain-computer interface(BCI)can be used to decode neurological activities, which has been widely applied in the control of external devices. However, two factors still hinder the improvement of classification accuracy and robustness, especially in multi-class tasks. First, existing algorithms are based on a single space (measuring or source space). They suffer from the holistic low spatial resolution of the measuring space or the locally high spatial resolution information accessed from the source space, failing to provide holistic and high-resolution representations. Second, the subject specificity is not sufficiently characterized, resulting in the loss of personalized intrinsic information. Therefore, we propose a cross-space convolutional neural network (CS-CNN) with customized characteristics for four-class MI-EEG classification. This algorithm uses the modified customized band common spatial patterns (CBCSP) and duplex mean-shift clustering (DMSClustering) to express the specific rhythms and source distribution information in cross-space. At the same time, multi-view features from the time, frequency and space domains are extracted, connecting with CNN to fuse the characteristics from two spaces and classify them. MI-EEG was collected from 20 subjects. Lastly, the classification accuracy of the proposed is 96.05% with real MRI information and 94.79% without MRI in the private dataset. And the results in the BCI competition IV-2a show that CS-CNN outperforms the state-of-the-art algorithms, achieving an accuracy improvement of 1.98%, and a standard deviation reduction of 5.15%.

A multi-feature fusion decoding study for unilateral upper-limb fine motor imagery.

To address the fact that the classical motor imagination paradigm has no noticeable effect on the rehabilitation training of upper limbs in patients after stroke and the corresponding feature extraction algorithm is limited to a single domain, this paper describes the design of a unilateral upper-limb fine motor imagination paradigm and the collection of data from 20 healthy people. It presents a feature extraction algorithm for multi-domain fusion and compares the common spatial pattern (CSP), improved multiscale permutation entropy (IMPE) and multi-domain fusion features of all participants through the use of decision tree, linear discriminant analysis, naive Bayes, a support vector machine, k-nearest neighbor and ensemble classification precision algorithms in the ensemble classifier. For the same subject, the average classification accuracy improvement of the same classifier for multi-domain feature extraction relative to CSP feature results went up by 1.52%. The average classification accuracy improvement of the same classifier went up by 32.87% relative to the IMPE feature classification results. This study's unilateral fine motor imagery paradigm and multi-domain feature fusion algorithm provide new ideas for upper limb rehabilitation after stroke.

Dual attention guided multiscale neural network trained with curriculum learning for noninvasive prediction of Gleason Grade Group from MRI.

BACKGROUND: The Gleason Grade Group (GG) is essential in assessing the malignancy of prostate cancer (PCa) and is typically obtained by invasive biopsy procedures in which sampling errors could lead to inaccurately scored GGs. With the gradually recognized value of bi-parametric magnetic resonance imaging (bpMRI) in PCa, it is beneficial to noninvasively predict GGs from bpMRI for early diagnosis and treatment planning of PCa. However, it is challenging to establish the connection between bpMRI features and GGs. PURPOSE: In this study, we propose a dual attention-guided multiscale neural network (DAMS-Net) to predict the 5-scored GG from bpMRI and design a training curriculum to further improve the prediction performance. METHODS: The proposed DAMS-Net incorporates a feature pyramid network (FPN) to fully extract the multiscale features for lesions of varying sizes and a dual attention module to focus on lesion and surrounding regions while avoiding the influence of irrelevant ones. Furthermore, to enhance the differential ability for lesions with the inter-grade similarity and intra-grade variation in bpMRI, the training process employs a specially designed curriculum based on the differences between the radiological evaluations and the ground truth GGs. RESULTS: Extensive experiments were conducted on a private dataset of 382 patients and the public PROSTATEx-2 dataset. For the private dataset, the experimental results showed that the proposed network performed better than the plain baseline model for GG prediction, achieving a mean quadratic weighted Kappa (Kw ) of 0.4902 and a mean positive predictive value of 0.9098 for predicting clinically significant cancer (PPVGG>1 ). With the application of curriculum learning, the mean Kw and PPVGG>1 further increased to 0.5144 and 0.9118, respectively. For the public dataset, the proposed method achieved state-of-the-art results of 0.5413 Kw and 0.9747 PPVGG>1 . CONCLUSION: The proposed DAMS-Net trained with curriculum learning can effectively predict GGs from bpMRI, which may assist clinicians in early diagnosis and treatment planning for PCa patients.

[Multi-task motor imagery electroencephalogram classification based on adaptive time-frequency common spatial pattern combined with convolutional neural network].

The effective classification of multi-task motor imagery electroencephalogram (EEG) is helpful to achieve accurate multi-dimensional human-computer interaction, and the high frequency domain specificity between subjects can improve the classification accuracy and robustness. Therefore, this paper proposed a multi-task EEG signal classification method based on adaptive time-frequency common spatial pattern (CSP) combined with convolutional neural network (CNN). The characteristics of subjects' personalized rhythm were extracted by adaptive spectrum awareness, and the spatial characteristics were calculated by using the one-versus-rest CSP, and then the composite time-domain characteristics were characterized to construct the spatial-temporal frequency multi-level fusion features. Finally, the CNN was used to perform high-precision and high-robust four-task classification. The algorithm in this paper was verified by the self-test dataset containing 10 subjects (33 ± 3 years old, inexperienced) and the dataset of the 4th 2018 Brain-Computer Interface Competition (BCI competition Ⅳ-2a). The average accuracy of the proposed algorithm for the four-task classification reached 93.96% and 84.04%, respectively. Compared with other advanced algorithms, the average classification accuracy of the proposed algorithm was significantly improved, and the accuracy range error between subjects was significantly reduced in the public dataset. The results show that the proposed algorithm has good performance in multi-task classification, and can effectively improve the classification accuracy and robustness.

Automated quantitative assessment of pediatric blunt hepatic trauma by deep learning-based CT volumetry.

BACKGROUND: To develop an end-to-end deep learning method for automated quantitative assessment of pediatric blunt hepatic trauma based on contrast-enhanced computed tomography (CT). METHODS: This retrospective study included 170 children with blunt hepatic trauma between May 1, 2015, and August 30, 2021, who had undergone contrast-enhanced CT. Both liver parenchyma and liver trauma regions were manually segmented from CT images. Two deep convolutional neural networks (CNNs) were trained on 118 cases between May 1, 2015, and December 31, 2019, for liver segmentation and liver trauma segmentation. Liver volume and trauma volume were automatically calculated based on the segmentation results, and the liver parenchymal disruption index (LPDI) was computed as the ratio of liver trauma volume to liver volume. The segmentation performance was tested on 52 cases between January 1, 2020, and August 30, 2021. Correlation analysis among the LPDI, trauma volume, and the American Association for the Surgery of Trauma (AAST) liver injury grade was performed using the Spearman rank correlation. The performance of severity assessment of pediatric blunt hepatic trauma based on the LPDI and trauma volume was evaluated using receiver operating characteristic (ROC) analysis. RESULTS: The Dice, precision, and recall of the developed deep learning framework were 94.75, 94.11, and 95.46% in segmenting the liver and 72.91, 72.40, and 76.80% in segmenting the trauma regions. The LPDI and trauma volume were significantly correlated with AAST grade (rho = 0.823 and rho = 0.831, respectively; p < 0.001 for both). The area under the ROC curve (AUC) values for the LPDI and trauma volume to distinguish between high-grade and low-grade pediatric blunt hepatic trauma were 0.942 (95% CI, 0.882-1.000) and 0.952 (95% CI, 0.895-1.000), respectively. CONCLUSIONS: The developed end-to-end deep learning method is able to automatically and accurately segment the liver and trauma regions from contrast-enhanced CT images. The automated LDPI and liver trauma volume can act as objective and quantitative indexes to supplement the current AAST grading of pediatric blunt hepatic trauma.

Detection of mild cognitive impairment in type 2 diabetes mellitus based on machine learning using privileged information.

Type 2 diabetes mellitus (T2DM) patients may develop into mild cognitive impairment (MCI) or even dementia. However, there is lack of reliable machine learning model for detection MCI in T2DM patients based on machine learning method. In addition, the brain network changes associated with MCI have not been studied. The aim of this study is to develop a machine learning based algorithm to help detect MCI in T2DM. There are 164 participants were included in this study. They were divided into T2DM-MCI (n = 56), T2DM-nonMCI (n = 49), and normal controls (n = 59) according to the neuropsychological evaluation. Functional connectivity of each participant was constructed based on resting-state magnetic resonance imaging (rs-fMRI). Feature selection was used to reduce the feature dimension. Then the selected features were set into the cascaded multi-column random vector functional link network (RVFL) classifier model using privileged information. Finally, the optimal model was trained and the classification performance was obtained using the testing data. The results show that the proposed algorithm has outstanding performance compared with classic methods. The classification accuracy of 73.18 % (T2DM-MCI vs NC) and 79.42 % (T2DM-MCI vs T2DM-nonMCI) were achieved. The functional connectivity related to T2DM-MCI mainly distribute in the frontal lobe, temporal lobe, and central region (motor cortex), which could be used as neuroimaging biomarkers to recognize MCI in T2DM patients. This study provides a machine learning model for diagnosis of MCI in T2DM patients and has potential clinical significance for timely intervention and treatment to delay the development of MCI.

Development of an ultrasound-based radiomics nomogram to preoperatively predict Ki-67 expression level in patients with breast cancer.

Objective: To develop and validate a radiomics nomogram that could incorporate clinicopathological characteristics and ultrasound (US)-based radiomics signature to non-invasively predict Ki-67 expression level in patients with breast cancer (BC) preoperatively. Methods: A total of 328 breast lesions from 324 patients with BC who were pathologically confirmed in our hospital from June 2019 to October 2020 were included, and they were divided into high Ki-67 expression level group and low Ki-67 expression level group. Routine US and shear wave elastography (SWE) were performed for each lesion, and the ipsilateral axillary lymph nodes (ALNs) were scanned for abnormal changes. The datasets were randomly divided into training and validation cohorts with a ratio of 7:3. Correlation analysis and the least absolute shrinkage and selection operator (LASSO) were used to select the radiomics features obtained from gray-scale US images of BC patients, and each radiomics score (Rad-score) was calculated. Afterwards, multivariate logistic regression analysis was used to establish a radiomics nomogram based on the radiomics signature and clinicopathological characteristics. The prediction performance of the nomogram was assessed by the area under the receiver operating characteristic curve (AUC), the calibration curve, and decision curve analysis (DCA) using the results of immunohistochemistry as the gold standard. Results: The radiomics signature, consisted of eight selected radiomics features, achieved a nearly moderate prediction efficacy with AUC of 0.821 (95% CI:0.764-0.880) and 0.713 (95% CI:0.612-0.814) in the training and validation cohorts, respectively. The radiomics nomogram, incorporating maximum diameter of lesions, stiff rim sign, US-reported ALN status, and radiomics signature showed a promising performance for prediction of Ki-67 expression level, with AUC of 0.904 (95% CI:0.860-0.948) and 0.890 (95% CI:0.817-0.964) in the training and validation cohorts, respectively. The calibration curve and DCA indicated promising consistency and clinical applicability. Conclusion: The proposed US-based radiomics nomogram could be used to non-invasively predict Ki-67 expression level in BC patients preoperatively, and to assist clinicians in making reliable clinical decisions.