Mayidili Nijiati
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Mayidili Nijiati has a total of 70 co-authors. Here are the top co-authors by number of shared publications:
Yasen Yimit
Medical Imaging Center, The First People’s Hospital of Kashi Prefecture, Kashi, People’s Republic of China
7 shared publicationsParhat Yasin
The Sixth Affiliated Hospital of Xinjiang Medical University Department of Spine Surgery, Urumqi, China
5 shared publicationsAbudoukeyoumujiang Abulizi
Department of Radiology, The First People’s Hospital of Kashi (Kashgar) Prefecture, Kashgar, China
4 shared publicationsChencui Huang
Department of Deepwise AI lab, Hangzhou Deepwise and League of PHD Technology Co., Ltd, Hangzhou, People’s Republic of China
4 shared publicationsAbudouresuli Tuersun
Department of Radiology, The First People’s Hospital of Kashi (Kashgar) Prefecture, Kashi, China
3 shared publicationsJuan Ma
Medical Imaging Center, Xinjiang Medical University Affiliated Fourth Hospital
3 shared publications
Latest Publications
- Open Access
- 25 June 2026
- Published by Frontiers Media SA in Frontiers in Physiology
- Vol. 17
Deep learning based automated HER2 score prediction using immunohistochemistry histopathological images: a dual-center study
Background: HER2 is a critical prognostic biomarker in breast cancer and associated with aggressive tumor biology. Current IHC scoring is subjective and labor-intensive. Deep learning has demonstrated success in histopathological image analysis, yet HER2 IHC automation remains underexplored. External-center validation is essential to establish clinical credibility and demonstrate robustness across diverse institutional practices and imaging protocols. Methods: This dual-center retrospective study analyzed 135 HER2 IHC whole-slide images from 118 breast cancer patients labeled as 1+, 2+, or 3+ by standard clinical criteria. Two board-certified pathologists manually annotated tumor-enriched ROIs, which were tiled into non-overlapping 512x512 patches; tiles with >60% white background were excluded. Patches were harmonized using a modified Macenko color normalization and augmented during training. Six pretrained deep learning models (AlexNet, VGG16, ResNet34, DenseNet121, Inception, Swin Transformer) were trained with patient-level splits and evaluated on an independent test set using macro-averaged AUC and complementary metrics. Results: The cohort included 118 patients with comparable age and largely similar baseline imaging/pathologic characteristics across groups, although clinical symptoms and lymph node status differed. On the independent test set, all models showed good discrimination for three-class HER2 grading, with AlexNet performing best (macro-AUC 0.971), followed by VGG16 (0.967). For AlexNet, per-class AUCs were 0.980 (1+), 0.955 (2+), and 0.979 (3+); most errors occurred between adjacent grades (1+/2+, 2+/3+). Grad-CAM highlighted strongly stained tumor regions driving predictions. Conclusion: A deep learning framework showed encouraging patch-level performance for three-class HER2 IHC score prediction in a pooled dual-center retrospective cohort. This approach may assist pathologists by improving scoring consistency and identifying borderline or low-confidence cases that require careful review and, when clinically indicated, confirmatory testing.
- 25 June 2026
- Published by Frontiers Media SA in Frontiers in Physiology
- Vol. 17
Open AccessDeep learning based automated HER2 score prediction using immunohistochemistry histopathological images: a dual-center study
Background: HER2 is a critical prognostic biomarker in breast cancer and associated with aggressive tumor biology. Current IHC scoring is subjective and labor-intensive. Deep learning has demonstrated success in histopathological image analysis, yet HER2 IHC automation remains underexplored. External-center validation is essential to establish clinical credibility and demonstrate robustness across diverse institutional practices and imaging protocols. Methods: This dual-center retrospective study analyzed 135 HER2 IHC whole-slide images from 118 breast cancer patients labeled as 1+, 2+, or 3+ by standard clinical criteria. Two board-certified pathologists manually annotated tumor-enriched ROIs, which were tiled into non-overlapping 512x512 patches; tiles with >60% white background were excluded. Patches were harmonized using a modified Macenko color normalization and augmented during training. Six pretrained deep learning models (AlexNet, VGG16, ResNet34, DenseNet121, Inception, Swin Transformer) were trained with patient-level splits and evaluated on an independent test set using macro-averaged AUC and complementary metrics. Results: The cohort included 118 patients with comparable age and largely similar baseline imaging/pathologic characteristics across groups, although clinical symptoms and lymph node status differed. On the independent test set, all models showed good discrimination for three-class HER2 grading, with AlexNet performing best (macro-AUC 0.971), followed by VGG16 (0.967). For AlexNet, per-class AUCs were 0.980 (1+), 0.955 (2+), and 0.979 (3+); most errors occurred between adjacent grades (1+/2+, 2+/3+). Grad-CAM highlighted strongly stained tumor regions driving predictions. Conclusion: A deep learning framework showed encouraging patch-level performance for three-class HER2 IHC score prediction in a pooled dual-center retrospective cohort. This approach may assist pathologists by improving scoring consistency and identifying borderline or low-confidence cases that require careful review and, when clinically indicated, confirmatory testing.
- Open Access
- 10 June 2026
- Published by Frontiers Media SA in Frontiers in Public Health
- Vol. 14, 1780946
Computer-aided detection of equivocal spinal tuberculosis on X-ray using a YOLOv11-based deep learning model
Background: Spinal tuberculosis (STB), also known as Pott's disease or tuberculous spondylitis, remains a substantial clinical burden in some remote and resource-limited regions, where X-ray imaging is often the primary modality available for initial assessment. However, early- and mid-stage STB frequently presents with subtle, equivocal, or atypical structural changes on X-ray images, which may lead to delayed recognition and referral in primary-care settings. Investigating suspicious-region localization in cases with clinical suspicion of STB but equivocal X-ray findings may provide supportive information for further imaging evaluation and referral decisions in primary-care settings. Methods: This retrospective study included 307 patients from three tertiary hospitals, all of whom had equivocal X-ray findings and CT-confirmed STB. CT-referenced suspicious-region annotations were established on X-ray images according to the CT-confirmed involved spinal levels and their anatomical correspondence on radiographs. A single-class suspicious-region localization model was developed based on the YOLOv11 object detection framework, and the dataset was split at the patient level into training and test sets. Model performance was evaluated using mAP@0.5, mAP@0.5:0.95, precision, recall, F1-confidence curves, precision-recall curves, and confidence-related curves. Patient-level detection rate and missed-case rate were further introduced to evaluate case-level alerting performance. To assess the spatial correspondence between CT-referenced annotations and suspicious regions visible on X-ray images, an agreement analysis between X-ray-only blinded annotations and CT-referenced annotations was additionally performed. Results: On the test set, the YOLOv11 model achieved an mAP@0.5 of 0.7664, an mAP@0.5:0.95 of 0.4520, a precision of 0.8358, and a recall of 0.6215. Threshold-related curves showed a trade-off between precision and recall across different confidence thresholds. At the provisional working threshold of 0.25, the patient-level detection rate was 0.8073, with a missed-case rate of 0.1927, indicating that the model could provide at least one suspicious-region prompt in most CT-confirmed positive test cases. The agreement analysis showed that all 40 randomly selected test cases could be annotated with at least one suspicious region based on X-ray images alone. The mean maximum IoU between X-ray-only blinded annotations and CT-referenced annotations was 0.7542; 34/40 cases reached a maxIoU ≥0.5, and 38/40 cases reached a maxIoU ≥0.3. Conclusion: This study shows that, using X-ray images alone as input, a deep learning model can provide suspicious-region localization prompts for CT-confirmed STB cases with equivocal X-ray findings. The proposed approach provides a feasible proof-of-concept framework for early risk prompting and referral support for STB in resource-limited settings. Further studies incorporating normal cases and non-tuberculous spinal disease controls, as well as external validation and prospective evaluation, are required to assess its applicability in real-world primary-care settings.
- 10 June 2026
- Published by Frontiers Media SA in Frontiers in Public Health
- Vol. 14, 1780946
Open AccessComputer-aided detection of equivocal spinal tuberculosis on X-ray using a YOLOv11-based deep learning model
Background: Spinal tuberculosis (STB), also known as Pott's disease or tuberculous spondylitis, remains a substantial clinical burden in some remote and resource-limited regions, where X-ray imaging is often the primary modality available for initial assessment. However, early- and mid-stage STB frequently presents with subtle, equivocal, or atypical structural changes on X-ray images, which may lead to delayed recognition and referral in primary-care settings. Investigating suspicious-region localization in cases with clinical suspicion of STB but equivocal X-ray findings may provide supportive information for further imaging evaluation and referral decisions in primary-care settings. Methods: This retrospective study included 307 patients from three tertiary hospitals, all of whom had equivocal X-ray findings and CT-confirmed STB. CT-referenced suspicious-region annotations were established on X-ray images according to the CT-confirmed involved spinal levels and their anatomical correspondence on radiographs. A single-class suspicious-region localization model was developed based on the YOLOv11 object detection framework, and the dataset was split at the patient level into training and test sets. Model performance was evaluated using mAP@0.5, mAP@0.5:0.95, precision, recall, F1-confidence curves, precision-recall curves, and confidence-related curves. Patient-level detection rate and missed-case rate were further introduced to evaluate case-level alerting performance. To assess the spatial correspondence between CT-referenced annotations and suspicious regions visible on X-ray images, an agreement analysis between X-ray-only blinded annotations and CT-referenced annotations was additionally performed. Results: On the test set, the YOLOv11 model achieved an mAP@0.5 of 0.7664, an mAP@0.5:0.95 of 0.4520, a precision of 0.8358, and a recall of 0.6215. Threshold-related curves showed a trade-off between precision and recall across different confidence thresholds. At the provisional working threshold of 0.25, the patient-level detection rate was 0.8073, with a missed-case rate of 0.1927, indicating that the model could provide at least one suspicious-region prompt in most CT-confirmed positive test cases. The agreement analysis showed that all 40 randomly selected test cases could be annotated with at least one suspicious region based on X-ray images alone. The mean maximum IoU between X-ray-only blinded annotations and CT-referenced annotations was 0.7542; 34/40 cases reached a maxIoU ≥0.5, and 38/40 cases reached a maxIoU ≥0.3. Conclusion: This study shows that, using X-ray images alone as input, a deep learning model can provide suspicious-region localization prompts for CT-confirmed STB cases with equivocal X-ray findings. The proposed approach provides a feasible proof-of-concept framework for early risk prompting and referral support for STB in resource-limited settings. Further studies incorporating normal cases and non-tuberculous spinal disease controls, as well as external validation and prospective evaluation, are required to assess its applicability in real-world primary-care settings.
- 22 April 2026
- Published by Springer Nature in The Visual Computer
- Vol. 42 (7) , 265
Enhancing representation learning with frequency-augmented feature mixture for robust tuberculosis screening
- 22 April 2026
- Published by Springer Nature in The Visual Computer
- Vol. 42 (7) , 265
Enhancing representation learning with frequency-augmented feature mixture for robust tuberculosis screening
- Open Access
- 22 November 2025
- Published by MDPI AG in Diagnostics
- Vol. 15 (23) , 2963
Multimodal-Imaging-Based Interpretable Deep Learning Framework for Distinguishing Brucella from Tuberculosis Spondylitis: A Dual-Center Study
Objectives: Brucella spondylitis (BS) and tuberculosis spondylitis (TS) are two causes of infection that share overlapping clinical and imaging features, complicating diagnoses. Early differentiation is critical, as treatment regimens differ significantly. This study aims to develop a deep learning framework using multimodal computed tomography (CT) and magnetic resonance imaging (MRI) data to accurately distinguish between these two conditions, improving diagnostic accuracy and patient outcomes. Methods: In this study, imaging data were acquired from two centers using different MRI and CT protocols. Sagittal T1-weighted (T1WI) and T2-weighted imaging (T2WI), fat-suppression sequences (T2WI FSE), and sagittal CT data were collected. Image preprocessing included region of interest (ROI) segmentation, and normalization and augmentation techniques were used. A deep learning model, based on pre-trained GoogleNet architectures, was trained and evaluated against human radiologists using metrics including accuracy, sensitivity, and AUC to assess diagnostic performance. Results: In this study, the GoogleNet deep learning model outperformed other architectures in classifying TS and BS, achieving AUCs of 95.97%, 91.24%, and 81.25% across training, test, and external validation datasets, respectively. In contrast, ResNet, DenseNet, and EfficientNet models showed lower AUC values. GoogleNet also demonstrated high accuracy (90.77% training, 83.04% test) and 90.91% sensitivity and 61.11% specificity in external validation. When compared to three radiologists, GoogleNet outperformed in diagnostic accuracy and speed, achieving an AUC of 88.01% and processing cases in 0.001 min. These findings highlight the potential of AI to enhance diagnostic performance and efficiency. Lastly, the explanation provided by the Grad-Cam model precisely localized major lesions. Conclusions: This multimodal-imaging-based deep learning model could well differentiate TS and BS. Deep learning does not need manual feature extraction, selection, or model development, and has great potential in daily clinical practice.
- 22 November 2025
- Published by MDPI AG in Diagnostics
- Vol. 15 (23) , 2963
Open AccessMultimodal-Imaging-Based Interpretable Deep Learning Framework for Distinguishing Brucella from Tuberculosis Spondylitis: A Dual-Center Study
Objectives: Brucella spondylitis (BS) and tuberculosis spondylitis (TS) are two causes of infection that share overlapping clinical and imaging features, complicating diagnoses. Early differentiation is critical, as treatment regimens differ significantly. This study aims to develop a deep learning framework using multimodal computed tomography (CT) and magnetic resonance imaging (MRI) data to accurately distinguish between these two conditions, improving diagnostic accuracy and patient outcomes. Methods: In this study, imaging data were acquired from two centers using different MRI and CT protocols. Sagittal T1-weighted (T1WI) and T2-weighted imaging (T2WI), fat-suppression sequences (T2WI FSE), and sagittal CT data were collected. Image preprocessing included region of interest (ROI) segmentation, and normalization and augmentation techniques were used. A deep learning model, based on pre-trained GoogleNet architectures, was trained and evaluated against human radiologists using metrics including accuracy, sensitivity, and AUC to assess diagnostic performance. Results: In this study, the GoogleNet deep learning model outperformed other architectures in classifying TS and BS, achieving AUCs of 95.97%, 91.24%, and 81.25% across training, test, and external validation datasets, respectively. In contrast, ResNet, DenseNet, and EfficientNet models showed lower AUC values. GoogleNet also demonstrated high accuracy (90.77% training, 83.04% test) and 90.91% sensitivity and 61.11% specificity in external validation. When compared to three radiologists, GoogleNet outperformed in diagnostic accuracy and speed, achieving an AUC of 88.01% and processing cases in 0.001 min. These findings highlight the potential of AI to enhance diagnostic performance and efficiency. Lastly, the explanation provided by the Grad-Cam model precisely localized major lesions. Conclusions: This multimodal-imaging-based deep learning model could well differentiate TS and BS. Deep learning does not need manual feature extraction, selection, or model development, and has great potential in daily clinical practice.
- Open Access
- 4 November 2025
- Published by Springer Nature in BMC Infectious Diseases
- Vol. 25 (1) , 1-13
Multimodal imaging-based interpretable radiomics for differentiating brucella and tuberculosis spondylitis: a two-center study
Spondylitis, particularly infectious forms caused by Mycobacterium tuberculosis and Brucella species, presents significant clinical challenges due to overlapping symptoms and diagnostic difficulties. Accurate differentiation is crucial for effective treatment, necessitating advanced imaging techniques and radiomics to enhance diagnostic precision and improve patient outcomes in cases of tuberculosis spondylitis (TS) and brucella spondylitis (BS). This retrospective cohort study included 195 patients diagnosed with TS or BS from January 2020 to December 2024 in center1, with an external validation cohort of 57 patients in center 2. Inclusion criteria consisted of relevant clinical symptoms for at least six months, positive serological tests, Magnetic Resonance Imaging (MRI) abnormalities, and complete medical records. Imaging was performed at two centers, employing standardized protocols for T1-weighted(T1WI) and T2-weighted(T2WI) and fat-suppression T2WI (FS T2WI) MRI. Region of Interest (ROI) segmentation and radiomics feature extraction were conducted using the Deepwise platform, yielding a total of 1049 Computed Tomography (CT) and 5829 MRI features. Nine predictive models were developed and validated through nested five-fold cross-validation, assessing performance metrics such as Area Under the Curve (AUC), sensitivity, and specificity. Statistical significance was set at p < 0.05. A total of 195 patients with 207 lesions were analyzed, comprising 116 patients of TS and 79 patients of BS. An external validation cohort included 57 patients with 60 lesions. Nine predictive models were developed using selected features: the CT model utilized 80 features, while T1WI, T2WI, and FS T2WI models employed 14, 33, and 32 features, respectively. Multi-modality models (CT, T1WI, T2WI and FS T2WI model) combined features from various sequences, achieving optimal performance with an AUC of 0.8136 in validation dataset. Model efficacy was validated through receiver operating characteristic (ROC) curve analysis, decision curve analyses, and calibration plots. Additionally, SHapley Additive exPlanations(SHAP) analysis was used to interpret model predictions, identifying key influential features, which are detailed in the supplementary materials. Our research indicates that multimodal imaging-based radiomics hold significant potential for the differential diagnosis of BS and TS.
- 4 November 2025
- Published by Springer Nature in BMC Infectious Diseases
- Vol. 25 (1) , 1-13
Open AccessMultimodal imaging-based interpretable radiomics for differentiating brucella and tuberculosis spondylitis: a two-center study
Spondylitis, particularly infectious forms caused by Mycobacterium tuberculosis and Brucella species, presents significant clinical challenges due to overlapping symptoms and diagnostic difficulties. Accurate differentiation is crucial for effective treatment, necessitating advanced imaging techniques and radiomics to enhance diagnostic precision and improve patient outcomes in cases of tuberculosis spondylitis (TS) and brucella spondylitis (BS). This retrospective cohort study included 195 patients diagnosed with TS or BS from January 2020 to December 2024 in center1, with an external validation cohort of 57 patients in center 2. Inclusion criteria consisted of relevant clinical symptoms for at least six months, positive serological tests, Magnetic Resonance Imaging (MRI) abnormalities, and complete medical records. Imaging was performed at two centers, employing standardized protocols for T1-weighted(T1WI) and T2-weighted(T2WI) and fat-suppression T2WI (FS T2WI) MRI. Region of Interest (ROI) segmentation and radiomics feature extraction were conducted using the Deepwise platform, yielding a total of 1049 Computed Tomography (CT) and 5829 MRI features. Nine predictive models were developed and validated through nested five-fold cross-validation, assessing performance metrics such as Area Under the Curve (AUC), sensitivity, and specificity. Statistical significance was set at p < 0.05. A total of 195 patients with 207 lesions were analyzed, comprising 116 patients of TS and 79 patients of BS. An external validation cohort included 57 patients with 60 lesions. Nine predictive models were developed using selected features: the CT model utilized 80 features, while T1WI, T2WI, and FS T2WI models employed 14, 33, and 32 features, respectively. Multi-modality models (CT, T1WI, T2WI and FS T2WI model) combined features from various sequences, achieving optimal performance with an AUC of 0.8136 in validation dataset. Model efficacy was validated through receiver operating characteristic (ROC) curve analysis, decision curve analyses, and calibration plots. Additionally, SHapley Additive exPlanations(SHAP) analysis was used to interpret model predictions, identifying key influential features, which are detailed in the supplementary materials. Our research indicates that multimodal imaging-based radiomics hold significant potential for the differential diagnosis of BS and TS.
- Open Access
- 31 July 2025
- Published by Frontiers Media SA in Frontiers in Physiology
- Vol. 16, 1635923
Efficacy of artificial intelligence-based FFR technology for coronary CTA stenosis detection in clinical management of coronary artery disease: a systematic review
Coronary computed tomography angiography (CCTA) integrated with artificial intelligence (AI) technology, particularly AI-based fractional flow reserve (FFR) assessment, has emerged as a crucial tool in the diagnosis and treatment of coronary artery disease (CAD). Recent advances in AI technology have demonstrated promising applications of AI-based FFR in detecting coronary stenosis through CCTA. Current evidence suggests that AI-FFR offers significant advantages in diagnostic accuracy and clinical utility, potentially enhancing the efficiency of CAD management. However, challenges persist in algorithm robustness, data heterogeneity, and clinical implementation. This review synthesizes recent developments in AI-based FFR technology for coronary stenosis detection via CCTA, focusing on AI-assisted quantitative coronary CTA (AI-QCT), deep learning algorithms, and their applications in three-dimensional coronary reconstruction and hemodynamic simulation. We analyze comparative diagnostic performance between AI-FFR and conventional methods, providing insights for future research directions and clinical applications.
- 31 July 2025
- Published by Frontiers Media SA in Frontiers in Physiology
- Vol. 16, 1635923
Open AccessEfficacy of artificial intelligence-based FFR technology for coronary CTA stenosis detection in clinical management of coronary artery disease: a systematic review
Coronary computed tomography angiography (CCTA) integrated with artificial intelligence (AI) technology, particularly AI-based fractional flow reserve (FFR) assessment, has emerged as a crucial tool in the diagnosis and treatment of coronary artery disease (CAD). Recent advances in AI technology have demonstrated promising applications of AI-based FFR in detecting coronary stenosis through CCTA. Current evidence suggests that AI-FFR offers significant advantages in diagnostic accuracy and clinical utility, potentially enhancing the efficiency of CAD management. However, challenges persist in algorithm robustness, data heterogeneity, and clinical implementation. This review synthesizes recent developments in AI-based FFR technology for coronary stenosis detection via CCTA, focusing on AI-assisted quantitative coronary CTA (AI-QCT), deep learning algorithms, and their applications in three-dimensional coronary reconstruction and hemodynamic simulation. We analyze comparative diagnostic performance between AI-FFR and conventional methods, providing insights for future research directions and clinical applications.
- Open Access
- 28 April 2025
- Published by Frontiers Media SA in Frontiers in Molecular Biosciences
- Vol. 12, 1570860
MRI-based deep learning with clinical and imaging features to differentiate medulloblastoma and ependymoma in children
Background: Medulloblastoma (MB) and ependymoma (EM) in children share similarities in terms of age group, tumor location, and clinical presentation, which makes it challenging to clinically diagnose and distinguish them.Purpose: The present study aims to explore the effectiveness of T2-weighted magnetic resonance imaging (MRI)-based deep learning (DL) combined with clinical imaging features for differentiating MB from EM.Methods: Axial T2-weighted MRI sequences obtained from 201 patients across three study centers were used for model training and testing. The regions of interest were manually delineated by an experienced neuroradiologist with supervision by a senior radiologist. We developed a DL classifier using a pretrained AlexNet architecture that was fine-tuned on our dataset. To mitigate class imbalance, we implemented data augmentation and employed K-fold cross-validation to enhance model generalizability. For patient classification, we used two voting strategies: hard voting strategy in which the majority prediction was selected from individual image slices; soft voting strategy in which the prediction scores were averaged across slices with a threshold of 0.5. Additionally, a multimodality fusion model was constructed by integrating the DL classifier with clinical and imaging features. The model performance was assessed using a 7:3 random split of the dataset for training and validation, respectively. The key metrics like sensitivity, specificity, positive predictive value, negative predictive value, F1 score, area under the receiver operating characteristic curve (AUC), and accuracy were calculated, and statistical comparisons were performed using the DeLong test. Thereafter, MB was classified as positive, while EM was classified as negative.Results: The DL model with the hard voting strategy achieved AUC values of 0.712 (95% confidence interval (CI): 0.625–0.797) on the training set and 0.689 (95% CI: 0.554–0.826) on the test set. In contrast, the multimodality fusion model demonstrated superior performance with AUC values of 0.987 (95% CI: 0.974–0.996) on the training set and 0.889 (95% CI: 0.803–0.949) on the test set. The DeLong test indicated a statistically significant improvement in AUC values for the fusion model compared to the DL model (p < 0.001), highlighting its enhanced discriminative ability.Conclusion: T2-weighted MRI-based DL combined with multimodal clinical and imaging features can be used to effectively differentiate MB from EM in children. Thus, the structure of the decision tree in the decision tree classifier is expected to greatly assist clinicians in daily practice.
- 28 April 2025
- Published by Frontiers Media SA in Frontiers in Molecular Biosciences
- Vol. 12, 1570860
Open AccessMRI-based deep learning with clinical and imaging features to differentiate medulloblastoma and ependymoma in children
Background: Medulloblastoma (MB) and ependymoma (EM) in children share similarities in terms of age group, tumor location, and clinical presentation, which makes it challenging to clinically diagnose and distinguish them.Purpose: The present study aims to explore the effectiveness of T2-weighted magnetic resonance imaging (MRI)-based deep learning (DL) combined with clinical imaging features for differentiating MB from EM.Methods: Axial T2-weighted MRI sequences obtained from 201 patients across three study centers were used for model training and testing. The regions of interest were manually delineated by an experienced neuroradiologist with supervision by a senior radiologist. We developed a DL classifier using a pretrained AlexNet architecture that was fine-tuned on our dataset. To mitigate class imbalance, we implemented data augmentation and employed K-fold cross-validation to enhance model generalizability. For patient classification, we used two voting strategies: hard voting strategy in which the majority prediction was selected from individual image slices; soft voting strategy in which the prediction scores were averaged across slices with a threshold of 0.5. Additionally, a multimodality fusion model was constructed by integrating the DL classifier with clinical and imaging features. The model performance was assessed using a 7:3 random split of the dataset for training and validation, respectively. The key metrics like sensitivity, specificity, positive predictive value, negative predictive value, F1 score, area under the receiver operating characteristic curve (AUC), and accuracy were calculated, and statistical comparisons were performed using the DeLong test. Thereafter, MB was classified as positive, while EM was classified as negative.Results: The DL model with the hard voting strategy achieved AUC values of 0.712 (95% confidence interval (CI): 0.625–0.797) on the training set and 0.689 (95% CI: 0.554–0.826) on the test set. In contrast, the multimodality fusion model demonstrated superior performance with AUC values of 0.987 (95% CI: 0.974–0.996) on the training set and 0.889 (95% CI: 0.803–0.949) on the test set. The DeLong test indicated a statistically significant improvement in AUC values for the fusion model compared to the DL model (p < 0.001), highlighting its enhanced discriminative ability.Conclusion: T2-weighted MRI-based DL combined with multimodal clinical and imaging features can be used to effectively differentiate MB from EM in children. Thus, the structure of the decision tree in the decision tree classifier is expected to greatly assist clinicians in daily practice.
- Open Access
- 28 August 2024
- Published by Springer Nature in BMC Infectious Diseases
- Vol. 24 (1) , 1-14
Predicting hospitalization costs for pulmonary tuberculosis patients based on machine learning
Pulmonary tuberculosis (PTB) is a prevalent chronic disease associated with a significant economic burden on patients. Using machine learning to predict hospitalization costs can allocate medical resources effectively and optimize the cost structure rationally, so as to control the hospitalization costs of patients better. This research analyzed data (2020–2022) from a Kashgar pulmonary hospital’s information system, involving 9570 eligible PTB patients. SPSS 26.0 was used for multiple regression analysis, while Python 3.7 was used for random forest regression (RFR) and MLP. The training set included data from 2020 and 2021, while the test set included data from 2022. The models predicted seven various costs related to PTB patients, including diagnostic cost, medical service cost, material cost, treatment cost, drug cost, other cost, and total hospitalization cost. The model’s predictive performance was evaluated using R-square (R2), Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE) metrics. Among the 9570 PTB patients included in the study, the median and quartile of total hospitalization cost were 13,150.45 (9891.34, 19,648.48) yuan. Nine factors, including age, marital status, admission condition, length of hospital stay, initial treatment, presence of other diseases, transfer, drug resistance, and admission department, significantly influenced hospitalization costs for PTB patients. Overall, MLP demonstrated superior performance in most cost predictions, outperforming RFR and multiple regression; The performance of RFR is between MLP and multiple regression; The predictive performance of multiple regression is the lowest, but it shows the best results for Other costs. The MLP can effectively leverage patient information and accurately predict various hospitalization costs, achieving a rationalized structure of hospitalization costs by adjusting higher-cost inpatient items and balancing different cost categories. The insights of this predictive model also hold relevance for research in other medical conditions.
- 28 August 2024
- Published by Springer Nature in BMC Infectious Diseases
- Vol. 24 (1) , 1-14
Open AccessPredicting hospitalization costs for pulmonary tuberculosis patients based on machine learning
Pulmonary tuberculosis (PTB) is a prevalent chronic disease associated with a significant economic burden on patients. Using machine learning to predict hospitalization costs can allocate medical resources effectively and optimize the cost structure rationally, so as to control the hospitalization costs of patients better. This research analyzed data (2020–2022) from a Kashgar pulmonary hospital’s information system, involving 9570 eligible PTB patients. SPSS 26.0 was used for multiple regression analysis, while Python 3.7 was used for random forest regression (RFR) and MLP. The training set included data from 2020 and 2021, while the test set included data from 2022. The models predicted seven various costs related to PTB patients, including diagnostic cost, medical service cost, material cost, treatment cost, drug cost, other cost, and total hospitalization cost. The model’s predictive performance was evaluated using R-square (R2), Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE) metrics. Among the 9570 PTB patients included in the study, the median and quartile of total hospitalization cost were 13,150.45 (9891.34, 19,648.48) yuan. Nine factors, including age, marital status, admission condition, length of hospital stay, initial treatment, presence of other diseases, transfer, drug resistance, and admission department, significantly influenced hospitalization costs for PTB patients. Overall, MLP demonstrated superior performance in most cost predictions, outperforming RFR and multiple regression; The performance of RFR is between MLP and multiple regression; The predictive performance of multiple regression is the lowest, but it shows the best results for Other costs. The MLP can effectively leverage patient information and accurately predict various hospitalization costs, achieving a rationalized structure of hospitalization costs by adjusting higher-cost inpatient items and balancing different cost categories. The insights of this predictive model also hold relevance for research in other medical conditions.
- Open Access
- 25 July 2024
- Published by Springer Nature in European Journal of Medical Research
- Vol. 29 (1) , 1-22
Machine learning-enabled prediction of prolonged length of stay in hospital after surgery for tuberculosis spondylitis patients with unbalanced data: a novel approach using explainable artificial intelligence (XAI)
Background: Tuberculosis spondylitis (TS), commonly known as Pott’s disease, is a severe type of skeletal tuberculosis that typically requires surgical treatment. However, this treatment option has led to an increase in healthcare costs due to prolonged hospital stays (PLOS). Therefore, identifying risk factors associated with extended PLOS is necessary. In this research, we intended to develop an interpretable machine learning model that could predict extended PLOS, which can provide valuable insights for treatments and a web-based application was implemented. Methods: We obtained patient data from the spine surgery department at our hospital. Extended postoperative length of stay (PLOS) refers to a hospitalization duration equal to or exceeding the 75th percentile following spine surgery. To identify relevant variables, we employed several approaches, such as the least absolute shrinkage and selection operator (LASSO), recursive feature elimination (RFE) based on support vector machine classification (SVC), correlation analysis, and permutation importance value. Several models using implemented and some of them are ensembled using soft voting techniques. Models were constructed using grid search with nested cross-validation. The performance of each algorithm was assessed through various metrics, including the AUC value (area under the curve of receiver operating characteristics) and the Brier Score. Model interpretation involved utilizing methods such as Shapley additive explanations (SHAP), the Gini Impurity Index, permutation importance, and local interpretable model-agnostic explanations (LIME). Furthermore, to facilitate the practical application of the model, a web-based interface was developed and deployed. Results: The study included a cohort of 580 patients and 11 features include (CRP, transfusions, infusion volume, blood loss, X-ray bone bridge, X-ray osteophyte, CT-vertebral destruction, CT-paravertebral abscess, MRI-paravertebral abscess, MRI-epidural abscess, postoperative drainage) were selected. Most of the classifiers showed better performance, where the XGBoost model has a higher AUC value (0.86) and lower Brier Score (0.126). The XGBoost model was chosen as the optimal model. The results obtained from the calibration and decision curve analysis (DCA) plots demonstrate that XGBoost has achieved promising performance. After conducting tenfold cross-validation, the XGBoost model demonstrated a mean AUC of 0.85 ± 0.09. SHAP and LIME were used to display the variables’ contributions to the predicted value. The stacked bar plots indicated that infusion volume was the primary contributor, as determined by Gini, permutation importance (PFI), and the LIME algorithm. Conclusions: Our methods not only effectively predicted extended PLOS but also identified risk factors that can be utilized for future treatments. The XGBoost model developed in this study is easily accessible through the deployed web application and can aid in clinical research.
- 25 July 2024
- Published by Springer Nature in European Journal of Medical Research
- Vol. 29 (1) , 1-22
Open AccessMachine learning-enabled prediction of prolonged length of stay in hospital after surgery for tuberculosis spondylitis patients with unbalanced data: a novel approach using explainable artificial intelligence (XAI)
Background: Tuberculosis spondylitis (TS), commonly known as Pott’s disease, is a severe type of skeletal tuberculosis that typically requires surgical treatment. However, this treatment option has led to an increase in healthcare costs due to prolonged hospital stays (PLOS). Therefore, identifying risk factors associated with extended PLOS is necessary. In this research, we intended to develop an interpretable machine learning model that could predict extended PLOS, which can provide valuable insights for treatments and a web-based application was implemented. Methods: We obtained patient data from the spine surgery department at our hospital. Extended postoperative length of stay (PLOS) refers to a hospitalization duration equal to or exceeding the 75th percentile following spine surgery. To identify relevant variables, we employed several approaches, such as the least absolute shrinkage and selection operator (LASSO), recursive feature elimination (RFE) based on support vector machine classification (SVC), correlation analysis, and permutation importance value. Several models using implemented and some of them are ensembled using soft voting techniques. Models were constructed using grid search with nested cross-validation. The performance of each algorithm was assessed through various metrics, including the AUC value (area under the curve of receiver operating characteristics) and the Brier Score. Model interpretation involved utilizing methods such as Shapley additive explanations (SHAP), the Gini Impurity Index, permutation importance, and local interpretable model-agnostic explanations (LIME). Furthermore, to facilitate the practical application of the model, a web-based interface was developed and deployed. Results: The study included a cohort of 580 patients and 11 features include (CRP, transfusions, infusion volume, blood loss, X-ray bone bridge, X-ray osteophyte, CT-vertebral destruction, CT-paravertebral abscess, MRI-paravertebral abscess, MRI-epidural abscess, postoperative drainage) were selected. Most of the classifiers showed better performance, where the XGBoost model has a higher AUC value (0.86) and lower Brier Score (0.126). The XGBoost model was chosen as the optimal model. The results obtained from the calibration and decision curve analysis (DCA) plots demonstrate that XGBoost has achieved promising performance. After conducting tenfold cross-validation, the XGBoost model demonstrated a mean AUC of 0.85 ± 0.09. SHAP and LIME were used to display the variables’ contributions to the predicted value. The stacked bar plots indicated that infusion volume was the primary contributor, as determined by Gini, permutation importance (PFI), and the LIME algorithm. Conclusions: Our methods not only effectively predicted extended PLOS but also identified risk factors that can be utilized for future treatments. The XGBoost model developed in this study is easily accessible through the deployed web application and can aid in clinical research.
- Open Access
- 20 March 2024
- Published by Elsevier in Academic Radiology
- Vol. 31 (8) , 3384-3396
Multiparametric MRI-Based Interpretable Radiomics Machine Learning Model Differentiates Medulloblastoma and Ependymoma in Children: A Two-Center Study
- 20 March 2024
- Published by Elsevier in Academic Radiology
- Vol. 31 (8) , 3384-3396
Open AccessMultiparametric MRI-Based Interpretable Radiomics Machine Learning Model Differentiates Medulloblastoma and Ependymoma in Children: A Two-Center Study