Independent Test Cohort Research Articles

Evaluation of cfDNA fragmentation characteristics in plasma for the diagnosis of lung cancer: A prospective cohort study.

Lung cancer is one of the most prevalent cancers worldwide, yet only approximately 16% of patients are diagnosed in early stage, highlighting the urgent need for novel, highly accurate detection models. In our study, patients with suspected lung cancer or lung disease, as identified through radiographic imaging, along with healthy individuals, were consecutively recruited from Beijing Chest Hospital. Circulating free DNA (cfDNA) was extracted from plasma samples, and low-depth whole-genome sequencing was performed to identify fragmentomic features for model construction. A total of 265 participants were prospectively enrolled, comprising 124 lung cancer patients and 141 noncancer individuals. The model we developed was based on four cfDNA fragmentation characteristics, including 20-bp breakpoint nucleotides motif, fragmentation size coverage, fragmentation size distribution, and copy number variation. In an independent test cohort, the model achieved an area under the curve (AUC) of 0.861 (95% CI: 0.781-0.942) and demonstrated a sensitivity of 70% (95% CI: 53.5%-83.4%) at a specificity of 89.4% (95% CI: 76.9%-96.5%). Notably, the model was also effective in detecting early-stage cancer, with an AUC of 0.808 (95% CI: 0.69-0.925). In summary, our lung cancer detection model shows strong screening capabilities by leveraging four cfDNA fragmentation characteristics, exhibiting robust performance in early cancer diagnosis.

Draw on advantages and avoid disadvantages: CT-derived individualized radiomic signature for predicting chemo-radiotherapy sensitivity in unresectable advanced non-small cell lung cancer

BackgroundPresently, the options of concurrent chemo-radiotherapy (CCR) in patients with locally advanced non-small cell lung cancer (LA-NSCLC) are controversial and there is no reliable prediction tool to stratify poor- and good-responders. Although radiomic analysis has provided new opportunities for personalized medicine in oncological practice, the repeatability and reproducibility of radiomic features are critical challenges that hinder their widespread clinical adoption. This study aimed to develop a qualitative radiomic signature based on the within-sample rank of radiomics features, and to use this novel method to predict CCR sensitivity in LA-NSCLC, avoiding the variability of quantitative signatures to multicenter effect.MethodsWe retrospectively analyzed 125 patients with stage III NSCLC who received treatment from our hospital. Radiomic features were extracted from pretreatment plain CT scans and constructed as feature pairs based on their within-sample rank. Fisher and univariate Cox analyses were performed to select feature pairs significantly associated with patients’ overall survival (OS). NSCLC-Radiomic (R422) cohort including 104 NSCLC patients was used as an independent testing cohort. NSCLC-Radiogenomic (RG211) cohort with matched RNA sequencing profiles, was used for functional enrichment analysis to reveal the underlying biological mechanism reflected by the signature.ResultsA qualitative signature, consisting of 15 radiomic feature pairs (termed as 15-RiFPS), was developed based on the Genetic Algorithm, which could optimally distinguish responder from non-responder with significantly improved OS if they received CCR treatment (log-rank P = 0.0009, HR = 13.79, 95% CIs 1.83–104.1). The performance of 15-RiFPS was validated in an independent public cohort (log-rank P = 0.0037, HR = 2.40, 95% CIs 1.30–4.40). Furthermore, the transcriptomic analyses provided biological pathways (‘glutathione metabolic process’, ‘cellular oxidant detoxification’) underlying the signature.ConclusionsWe developed a CT-derived 15-RiFPS, which could potentially help predict individualized therapeutic benefit of CCR in patients with LA-NSCLC. Additionally, we investigated the underlying intra-tumoral biological characteristics behind 15-RiFPS which would accelerate its clinical application. This approach could be applied to a wider range of treatments and cancer types.

Structure-function coupling in white matter uncovers the hypoconnectivity in autism spectrum disorder

BackgroundAutism Spectrum Disorder (ASD) is a neurodevelopmental disorder associated with alterations in structural and functional coupling in gray matter. However, despite the detectability and modulation of brain signals in white matter, the structure-function coupling in white matter in autism remains less explored.MethodsIn this study, we investigated structural-functional coupling in white matter (WM) regions, by integrating diffusion tensor data that contain fiber orientation information from WM tracts, with functional connectivity tensor data that reflect local functional anisotropy information. Using functional and diffusion magnetic resonance images, we analyzed a cohort of 89 ASD and 63 typically developing (TD) individuals from the Autism Brain Imaging Data Exchange II (ABIDE-II). Subsequently, the associations between structural-functional coupling in WM regions and ASD severity symptoms assessed by Autism Diagnostic Observation Schedule-2 were examined via supervised machine learning in an independent test cohort of 29 ASD individuals. Furthermore, we also compared the performance of multi-model features (i.e. structural-functional coupling) with single-model features (i.e. functional or structural models alone).ResultsIn the discovery cohort (ABIDE-II), individuals with ASD exhibited widespread reductions in structural-functional coupling in WM regions compared to TD individuals, particularly in commissural tracts (e.g. corpus callosum), association tracts (sagittal stratum), and projection tracts (e.g. internal capsule). Notably, supervised machine learning analysis in the independent test cohort revealed a significant correlation between these alterations in structural-functional coupling and ASD severity scores. Furthermore, compared to single-model features, the integration of multi-model features (i.e., structural-functional coupling) performed best in predicting ASD severity scores.ConclusionThis work provides novel evidence for atypical structural-functional coupling in ASD in white matter regions, further refining our understanding of the critical role of WM networks in the pathophysiology of ASD.

Development of a COVID-19 early risk assessment system based on multiple machine learning algorithms and routine blood tests: a real-world study.

During the Coronavirus Disease 2019 (COVID-19) epidemic, the massive spread of the disease has placed an enormous burden on the world's healthcare and economy. The early risk assessment system based on a variety of machine learning (ML) algorithms may be able to provide more accurate advice on the classification of COVID-19 patients, offering predictive, preventive, and personalized medicine (PPPM) solutions in the future. In this retrospective study, we divided a portion of the data into training and validation cohorts in a 7:3 ratio and established a model based on a combination of two ML algorithms first. Then, we used another portion of the data as an independent testing cohort to determine the most accurate and stable model and compared it with other scoring systems. Finally, patients were categorized according to risk scores and then the correlation between their clinical data and risk scores was studied. The elderly accounted for the majority of hospitalized patients with COVID-19. The C-index of the model constructed by combining the stepcox[both] and survivalSVM algorithms was 0.840 in the training cohort and 0.815 in the validation cohort, which was calculated to have the highest C-index in the testing cohort compared to the other 119 ML model combinations. Compared with current scoring systems, including the CURB-65 and several reported prognosis models previously, our model had the highest AUC value of 0.778, representing an even higher predictive performance. In addition, the model's AUC values for specific time intervals, including days 7,14 and 28, demonstrate excellent predictive performance. Most importantly, we stratified patients according to the model's risk score and demonstrated a difference in survival status between the high-risk, median-risk, and low-risk groups, which means a new and stable risk assessment system was built. Finally, we found that COVID-19 patients with a history of cerebral infarction had a significantly higher risk of death. This novel risk assessment system is highly accurate in predicting the prognosis of patients with COVID-19, especially elderly patients with COVID-19, and can be well applied within the PPPM framework. Our ML model facilitates stratified patient management, meanwhile promoting the optimal use of healthcare resources.

Advancing microvascular invasion diagnosis: a multi-center investigation of novel MRI-based models for precise detection and classification in early-stage small hepatocellular carcinoma (≤ 3cm).

This study aimed to develop two preoperative magnetic resonance imaging (MRI) based models for detecting and classifying microvascular invasion (MVI) in early-stage small hepatocellular carcinoma (sHCC) patients. MVI is graded as M0 (no invasion), M1 (invasion of five or fewer vessels located within 1cm of the tumor's peritumoral region), and M2 (invasion of more than five vessels or those located more than 1cm from the tumor's surface). This study enrolled 395 early-stage sHCC (≤ 3cm) patients from three centers who underwent preoperative gadopentetate-enhanced MRI. From the first two centers, 310 patients were randomly divided into training (n = 217) and validation (n = 93) cohorts in a 7:3 ratio to develop the first model for predicting MVI presence. Among these, 153 patients with identified MVI were further divided into training (n = 112) and validation (n = 41) cohorts, using the same ratio, to construct the second model for MVI classification. An independent test cohort of 85 patients from the third center to validate both models. Univariate and multivariate logistic regression analyses identified independent predictors of MVI and its classification in the training cohorts. Based on these predictors, two nomograms were developed and assessed for their discriminative ability, calibration, and clinical usefulness. Besides, considering the two models are supposed applied in a serial fashion in the real clinical setting, we evaluate the performance of the two models together on the test cohorts by applying them simultaneously. Kaplan-Meier survival curve analysis was employed to assess the correlation between predicted MVI status and early recurrence, similar to the association observed with actual MVI status and early recurrence. The MVI detection nomogram, with platelet count (PLT), activated partial thromboplastin time (APTT), rim arterial phase hyperenhancement (Rim APHE) and arterial peritumoral enhancement, achieved area under the curve (AUC) of 0.827, 0.761 and 0.798 in the training, validation, and test cohorts, respectively. The MVI classification nomogram, integrating Total protein (TP), Shape, Arterial peritumoral enhancement and enhancement pattern, achieved AUC of 0.824, 0.772, and 0.807 across the three cohorts. When the two models were applied on the test cohorts in a serial fashion, they both demonstrated good performance, which means the two models had good clinical applicability. Calibration and decision curve analysis (DCA) results affirmed the model's reliability and clinical utility. Notably, early recurrence was more prevalent in the MVI grade 2 (M2) group compared to the MVI-absent and M1 groups, regardless of the actual or predicted MVI status. The nomograms exhibited excellent predictive performance for detecting and classifying MVI in patients with early-stage sHCC, particularly identifying high-risk M2 patients preoperatively.

Machine learning-based prediction of early neurological deterioration after intravenous thrombolysis for stroke: insights from a large multicenter study.

This investigation seeks to ascertain the efficacy of various machine learning models in forecasting early neurological deterioration (END) following thrombolysis in patients with acute ischemic stroke (AIS). Employing data from the Shenyang Stroke Emergency Map database, this multicenter study compiled information on 7,570 AIS patients from 29 comprehensive hospitals who received thrombolytic therapy between January 2019 and December 2021. An independent testing cohort was constituted from 2,046 patients at the First People's Hospital of Shenyang. The dataset incorporated 15 pertinent clinical and therapeutic variables. The principal outcome assessed was the occurrence of END post-thrombolysis. Model development was executed using an 80/20 split for training and internal validation, employing classifiers like logistic regression with lasso regularization (lasso regression), support vector machine (SVM), random forest (RF), gradient-boosted decision tree (GBDT), and multi-layer perceptron (MLP). The model with the highest area under the curve (AUC) was utilized to delineate feature significance. Baseline characteristics showed variability in END incidence between the training (n = 7,570; END incidence 22%) and external validation cohorts (n = 2,046; END incidence 10%; p < 0.001). Notably, all machine learning models demonstrated superior AUC values compared to the reference model, indicating their enhanced predictive capacity. The lasso regression model achieved the highest AUC at 0.829 (95% CI: 0.799-0.86; p < 0.001), closely followed by the MLP model with an AUC of 0.828 (95% CI: 0.799-0.858; p < 0.001). The SVM, RF, and GBDT models also showed commendable AUCs of 0.753, 0.797, and 0.774, respectively. Decision curve analysis revealed that the SVM and MLP models demonstrated a high net benefit. Feature importance analysis emphasized "Onset To Needle Time" and "Admission NIHSS Score" as significant predictors. Our research establishes the MLP and lasso regression as robust tools for predicting early neurological deterioration in acute ischemic stroke patients following thrombolysis. Their superior predictive accuracy, compared to traditional models, highlights the significant potential of machine learning approaches in refining prognosis and enhancing clinical decisions in stroke care management. This advancement paves the way for more tailored therapeutic strategies, ultimately aiming to improve patient outcomes in clinical practice.

Clear Cell Renal Cell Carcinoma: Characterizing the Phenotype of Von Hippel-Lindau Mutation Using MRI.

The von Hippel-Lindau (VHL) mutation is an important alteration in clear cell renal cell carcinoma (ccRCC); however, its imaging phenotype remains unclear. To investigate whether MRI features can reflect the VHL mutation status. Retrospective. 3 T/fast spin echo T2-weighted, spin-echo echo planar diffusion-weighted, gradient recalled echo T1-weighted, gradient recalled echo chemical-shift T1-weighted, and contrast-enhanced gradient recalled echo T1-weighted sequences. One hundred five patients with ccRCC who underwent preoperative contrast-enhanced MRI and subsequent genomic sequencing: 59 consecutive patients from our institution (38 [64.41%] with VHL mutations) formed a training cohort, and 46 from The Cancer Genome Atlas (TCGA) database (24 [52.17%] with VHL mutations) formed an independent test cohort. Two radiologists, with 23 and 33 years of experience respectively, jointly evaluated the semantic MRI features of the primary lesion in ccRCCs to propose potential features related to VHL mutations in both cohorts. Three additional readers, with 5, 7, and 10 years of experience respectively, independently reviewed all lesions to assess the interobserver agreement of MRI features. A VHL mutational likelihood score (VHL-MULIS) system was constructed using the training cohort and validated using the independent test cohort. Fisher's test or chi-square test, t-test or Mann-Whitney U test, logistic regression, Cohen's kappa (κ), area under the receiver operating characteristic curve (AUC). A two-sided P value <0.05 was considered statistically significant. In both the local and public cohorts, T2-weighted signal intensity and presence of microscopic fat from primary lesions were significantly associated with VHL mutation status. The VHL-MULIS incorporated maximum diameter, T2-weighted signal intensity, and presence of microscopic fat in the training cohort and demonstrated promising diagnostic ability (AUC, 0.82; sensitivity, 0.79; specificity, 0.82) and substantial interobserver agreement (κ, 0.787) in the test cohort. The VHL mutation exhibited a distinct MRI phenotype. Integrating multiple semantic MRI features has potential to reflect the mutation status in patients with ccRCC. 3 TECHNICAL EFFICACY: Stage 2.

Fast intraoperative detection of primary CNS lymphoma and differentiation from common CNS tumors using stimulated Raman histology and deep learning.

Accurate intraoperative diagnosis is crucial for differentiating between primary CNS lymphoma (PCNSL) and other CNS entities, guiding surgical decision-making, but represents significant challenges due to overlapping histomorphological features, time constraints, and differing treatment strategies. We combined stimulated Raman histology (SRH) with deep learning to address this challenge. We imaged unprocessed, label-free tissue samples intraoperatively using a portable Raman scattering microscope, generating virtual H&E-like images within less than three minutes. We developed a deep learning pipeline called RapidLymphoma based on a self-supervised learning strategy to (1) detect PCNSL, (2) differentiate from other CNS entities, and (3) test the diagnostic performance in a prospective international multicenter cohort and two additional independent test cohorts. We trained on 54,000 SRH patch images sourced from surgical resections and stereotactic-guided biopsies, including various CNS tumor/non-tumor lesions. Training and test data were collected from four tertiary international medical centers. The final histopathological diagnosis served as ground-truth. In the prospective test cohort of PCNSL and non-PCNSL entities (n=160), RapidLymphoma achieved an overall balanced accuracy of 97.81% ±0.91, non-inferior to frozen section analysis in detecting PCNSL (100% vs. 78.94%). The additional test cohorts (n=420, n=59) reached balanced accuracy rates of 95.44% ±0.74 and 95.57% ±2.47 in differentiating IDH-wildtype diffuse gliomas and various brain metastasis from PCNSL. Visual heatmaps revealed RapidLymphoma's capabilities to detect class-specific histomorphological key features. RapidLymphoma is valid and reliable in detecting PCNSL and differentiating from other CNS entities within three minutes, as well as visual feedback in an intraoperative setting. This leads to fast clinical decision-making and further treatment strategy planning.

Endometrial cancer risk stratification using MRI radiomics: corroborating with choline metabolism

Background and purposeRadiomics offers little explainability. This study aims to develop a radiomics model (Rad-Score) using diffusion-weighted imaging (DWI) to predict high-risk patients for nodal metastasis or recurrence in endometrial cancer (EC) and corroborate with choline metabolism.Materials and methodsFrom August 2015 to July 2018, 356 EC patients were enrolled. Rad-Score was developed using LASSO regression in a training cohort (n = 287) and validated in an independent test cohort (n = 69). MR spectroscopy (MRS) was also used in 230 patients. Nuclear MRS measured choline metabolites in 70 tissue samples. The performance was compared against European Society for Medical Oncology (ESMO) risk groups. A P < .05 denoted statistical significance.ResultsRad-Score achieved 71.1% accuracy in the training and 71.0% in the testing cohorts. Incorporating clinical parameters of age, tumor type, size, and grade, Rad-Signature reached accuracies of 73.2% in training and 75.4% in testing cohorts, closely matching the performance to the post-operatively based ESMO's 70.7% and 78.3%. Rad-Score was significantly associated with increased total choline levels on MRS (P = .034) and tissue levels (P = .019).ConclusionsDevelopment of a preoperative radiomics risk score, comparable to ESMO clinical standard and associated with altered choline metabolism, shows translational relevance for radiomics in high-risk EC patients.Trial registrationThis study was registered in ClinicalTrials.gov on 2015–08-01 with Identifier NCT02528864.

Preoperative Ultrasound Radomics to Predict Posthepatectomy Liver Failure in Patients With Hepatocellular Carcinoma.

Posthepatectomy liver failure (PHLF) is a major cause of postoperative mortality in hepatocellular carcinoma (HCC) patients. The study aimed to develop a method based on the two-dimensional shear wave elastography and clinical data to evaluate the risk of PHLF in HCC patients with chronic hepatitis B. This multicenter study proposed a deep learning model (PHLF-Net) incorporating dual-modal ultrasound features and clinical indicators to predict the PHLF risk. The datasets were divided into a training cohort, an internal validation cohort, an internal independent testing cohort, and three external independent testing cohorts. Based on ResNet50 pretrained on ImageNet, PHLF-Net used a progressive training strategy with images of varying granularity and incorporated conventional B-mode and elastography images and clinical indicators related to liver reserve function. In total, 532 HCC patients who underwent hepatectomy at five hospitals were enrolled. PHLF occurred in 147 patients (27.6%, 147/532). The PHLF-Net combining dual-modal ultrasound and clinical indicators demonstrated high effectiveness for predicting PHLF, with AUCs of 0.957 and 0.923 in the internal validation and testing sets, and AUCs of 0.950, 0.860, and 1.000 in the other three independent external testing sets. The performance of PHLF-Net outperformed models of single- and dual-modal US. Preoperative ultrasound imaging combining clinical indicators can effectively predict the PHLF probability in patients with HCC. In the internal and external validation sets, PHLF-Net demonstrated its usefulness in predicting PHLF.

Integrating a deep neural network and Transformer architecture for the automatic segmentation and survival prediction in cervical cancer.

Automated tumor segmentation and survival prediction are critical to clinical diagnosis and treatment. This study aimed to develop deep-learning models for automatic tumor segmentation and survival prediction in magnetic resonance imaging (MRI) of cervical cancer (CC) by combining deep neural networks and Transformer architecture. This study included 406 patients with CC, each with comprehensive clinical information and MRI scans. We randomly divided patients into training, validation, and independent test cohorts in a 6:2:2 ratio. During the model training, we employed two architecture types: one being a hybrid model combining convolutional neural network (CNN) and ransformer (CoTr) and one of pure CNNs. For survival prediction, the hybrid model combined tumor image features extracted by segmentation models with clinical information. The performance of the segmentation models was evaluated using the Dice similarity coefficient (DSC) and 95% Hausdorff distance (HD95). The performance of the survival models was assessed using the concordance index. The CoTr model performed well in both contrast-enhanced T1-weighted (ceT1W) and T2-weighted (T2W) imaging segmentation tasks, with average DSCs of 0.827 and 0.820, respectively, which outperformed other the CNN models such as U-Net (DSC: 0.807 and 0.808), attention U-Net (DSC: 0.814 and 0.811), and V-Net (DSC: 0.805 and 0.807). For survival prediction, the proposed deep-learning model significantly outperformed traditional methods, yielding a concordance index of 0.732. Moreover, it effectively divided patients into low-risk and high-risk groups for disease progression (P<0.001). Combining Transformer architecture with a CNN can improve MRI tumor segmentation, and this deep-learning model excelled in the survival prediction of patients with CC as compared to traditional methods.

Single- and multi-site radiomics may improve overall survival prediction for patients with metastatic lung adenocarcinoma

PurposeThe purpose of this study was to assess whether single-site and multi-site radiomics could improve the prediction of overall survival (OS) of patients with metastatic lung adenocarcinoma compared to clinicopathological model. Materials and methodsAdults with metastatic lung adenocarcinoma, pretreatment whole-body contrast-enhanced computed tomography examinations, and performance status (WHO-PS) ≤ 2 were included in this retrospective single-center study, and randomly assigned to training and testing cohorts. Radiomics features (RFs) were extracted from all measurable lesions with volume ≥ 1 cm3. Radiomics prognostic scores based on the largest tumor (RPSlargest) and the average RF values across all tumors per patient (RPSaverage) were developed in the training cohort using 5-fold cross-validated LASSO-penalized Cox regression. Intra-patient inter-tumor heterogeneity (IPITH) metrics were calculated to quantify the radiophenotypic dissimilarities among all tumors within each patient. A clinicopathological model was built in the training cohort using stepwise Cox regression and enriched with combinations of RPSaverage, RPSlargest and IPITH. Models were compared with the concordance index in the independent testing cohort. ResultsA total of 300 patients (median age: 63.7 years; 40.7% women; median OS, 16.3 months) with 1359 lesions were included (200 and 100 patients in the training and testing cohorts, respectively). The clinicopathological model included WHO-PS = 2 (hazard ratio [HR] = 3.26; P < 0.0001), EGFR, ALK, ROS1 or RET mutations (HR = 0.57; P = 0.0347), IVB stage (HR = 1.65; P = 0.0211), and liver metastases (HR = 1.47; P = 0.0670). In the testing cohort, RPSaverage, RPSlargest and IPITH were associated with OS (HR = 85.50, P = 0.0038; HR = 18.83, P = 0.0082 and HR = 8.00, P = 0.0327, respectively). The highest concordance index was achieved with the combination of clinicopathological variables and RPSaverage, significantly better than that of the clinicopathological model (concordance index = 0.7150 vs. 0.695, respectively; P = 0.0049) ConclusionSingle-site and multi-site radiomics-based scores are associated with OS in patients with metastatic lung adenocarcinoma. RPSaverage improves the clinicopathological model.

Deep learning-based automatic ASPECTS calculation can improve diagnosis efficiency in patients with acute ischemic stroke: a multicenter study.

The Alberta Stroke Program Early CT Score (ASPECTS), a systematic method for assessing ischemic changes in acute ischemic stroke using non-contrast computed tomography (NCCT), is often interpreted relying on expert experience and can vary between readers. This study aimed to develop a clinically applicable automatic ASPECTS system employing deep learning (DL). This study enrolled 1987 NCCT scans that were retrospectively collected from four centers between January 2017 and October 2021. A DL-based system for automated ASPECTS assessment was trained on a development cohort (N = 1767) and validated on an independent test cohort (N = 220). The consensus of experienced physicians was regarded as a reference standard. The validity and reliability of the proposed system were assessed against physicians' readings. A real-world prospective application study with 13,399 patients was used for system validation in clinical contexts. The DL-based system achieved an area under the receiver operating characteristic curve (AUC) of 84.97% and an intraclass correlation coefficient (ICC) of 0.84 for overall-level analysis on the test cohort. The system's diagnostic sensitivity was 94.61% for patients with dichotomized ASPECTS at a threshold of ≥ 6, with substantial agreement (ICC = 0.65) with expert ratings. Combining the system with physicians improved AUC from 67.43 to 89.76%, reducing diagnosis time from 130.6 ± 66.3 s to 33.3 ± 8.3 s (p < 0.001). During the application in clinical contexts, 94.0% (12,591) of scans successfully processed by the system were utilized by clinicians, and 96% of physicians acknowledged significant improvement in work efficiency. The proposed DL-based system could accurately and rapidly determine ASPECTS, which might facilitate clinical workflow for early intervention. The deep learning-based automated ASPECTS evaluation system can accurately and rapidly determine ASPECTS for early intervention in clinical workflows, reducing processing time for physicians by 74.8%, but still requires validation by physicians when in clinical applications. The deep learning-based system for ASPECTS quantification has been shown to be non-inferior to expert-rated ASPECTS. This system improved the consistency of ASPECTS evaluation and reduced processing time to 33.3 seconds per scan. 94.0% of scans successfully processed by the system were utilized by clinicians during the prospective clinical application.

Comprehensive analysis of clinical data and radiomic features from contrast enhanced CT for differentiating benign and malignant pancreatic intraductal papillary mucinous neoplasms

The primary aim of this investigation was to leverage radiomics features derived from contrast-enhanced abdominal computed tomography (CT) scans to devise a predictive model to discern the benign and malignant nature of intraductal papillary mucinous neoplasms (IPMNs). Radiomic signatures were meticulously crafted to delineate benign from malignant IPMNs by extracting pertinent features from contrast-enhanced CT images within a designated training cohort (n = 84). Subsequent validation was conducted with data from an independent test cohort (n = 37). The discriminative ability of the model was quantitatively evaluated through receiver operating characteristic (ROC) curve analysis, with the integration of carefully selected clinical features to improve the comparative analysis. Arterial-phase images were utilized to construct a model comprising 8 features for distinguishing between benign and malignant cases. The model achieved an accuracy of 0.891 [95% confidence interval (95% CI), 0.816–0.996] in the cross-validation set and 0.553 (95% CI 0.360–0.745) in the test set. Conversely, employing 9 features from the venous-phase resulted in a model with a cross-validation accuracy of 0.862 (95%CI 0.777–0.946) and a test set accuracy of 0.801 (95% CI 0.653–0.950).Integrating the identified clinical features with imaging features yielded a model with a cross-validation accuracy of 0.934 (95% CI 0.879–0.990) and a test set accuracy of 0.904 (95% CI 0.808–0.999), thereby further improving its discriminatory ability. Our findings distinctly illustrate that venous-phase radiomics features eclipse arterial-phase radiomic features in terms of predictive accuracy regarding the nature of IPMNs. Furthermore, the synthesis and meticulous screening of clinical features with radiomic data significantly increased the diagnostic efficacy of our model, underscoring the pivotal importance of a comprehensive and integrated approach for accurate risk stratification in IPMN management.

Prediction of prognosis in lung cancer using machine learning with inter-institutional generalizability: A multicenter cohort study (WJOG15121L: REAL-WIND)

ObjectivesPredicting the prognosis of lung cancer is crucial for providing optimal medical care. However, a method to accurately predict the overall prognosis in patients with stage IV lung cancer, even with the use of machine learning, has not been established. Moreover, the inter-institutional generalizability of such algorithms remains unexplored. This study aimed to establish machine learning-based algorithms with inter-institutional generalizability to predict prognosis. Materials and MethodsThis multicenter, retrospective, hospital-based cohort study included consecutive patients with stage IV lung cancer who were randomly categorized into the training and independent test cohorts with a 2:1 ratio, respectively. The primary metric to assess algorithm performance was the area under the receiver operating characteristic curve in the independent test cohort. To assess the inter-institutional generalizability of the algorithms, we investigated their ability to predict patient outcomes in the remaining facility after being trained using data from 15 other facilities. ResultsOverall, 6,751 patients (median age, 70 years) were enrolled, and 1,515 (22 %) showed mutated epidermal growth factor receptor expression. The median overall survival was 16.6 (95 % confidence interval, 15.9–17.5) months. Algorithm performance metrics in the test cohort showed that the areas under the curves were 0.90 (95 % confidence interval, 0.88–0.91), 0.85 (0.84–0.87), 0.83 (0.81–0.85), and 0.85 (0.82–0.87) at 180, 360, 720, and 1,080 predicted survival days, respectively. The performance test of 16 algorithms for investigating inter-institutional generalizability showed median areas under the curves of 0.87 (range, 0.84–0.92), 0.84 (0.78–0.88), 0.84 (0.76–0.89), and 0.84 (0.75–0.90) at 180, 360, 720, and 1,080 days, respectively. ConclusionThis study developed machine learning algorithms that could accurately predict the prognosis in patients with stage IV lung cancer with high inter-institutional generalizability. This can enhance the accuracy of prognosis prediction and support informed and shared decision-making in clinical settings.

Quantifying impairment and disease severity using AI models trained on healthy subjects

Automatic assessment of impairment and disease severity is a key challenge in data-driven medicine. We propose a framework to address this challenge, which leverages AI models trained exclusively on healthy individuals. The COnfidence-Based chaRacterization of Anomalies (COBRA) score exploits the decrease in confidence of these models when presented with impaired or diseased patients to quantify their deviation from the healthy population. We applied the COBRA score to address a key limitation of current clinical evaluation of upper-body impairment in stroke patients. The gold-standard Fugl-Meyer Assessment (FMA) requires in-person administration by a trained assessor for 30-45 minutes, which restricts monitoring frequency and precludes physicians from adapting rehabilitation protocols to the progress of each patient. The COBRA score, computed automatically in under one minute, is shown to be strongly correlated with the FMA on an independent test cohort for two different data modalities: wearable sensors (ρ = 0.814, 95% CI [0.700,0.888]) and video (ρ = 0.736, 95% C.I [0.584, 0.838]). To demonstrate the generalizability of the approach to other conditions, the COBRA score was also applied to quantify severity of knee osteoarthritis from magnetic-resonance imaging scans, again achieving significant correlation with an independent clinical assessment (ρ = 0.644, 95% C.I [0.585,0.696]).

MRI radiomics enhances radiologists’ ability for characterizing intestinal fibrosis in patients with Crohn’s disease

ObjectivesWe aimed to develop MRI-based radiomic models (RMs) to improve the diagnostic accuracy of radiologists in characterizing intestinal fibrosis in patients with Crohn’s disease (CD).MethodsThis retrospective study included patients with refractory CD who underwent MR before surgery from November 2013 to September 2021. Resected bowel segments were histologically classified as none-mild or moderate-severe fibrosis. RMs based on different MR sequence combinations (RM1: T2WI and enhanced-T1WI; RM2: T2WI, enhanced-T1WI, diffusion-weighted imaging [DWI], and apparent diffusion coefficient [ADC]); RM3: T2WI, enhanced-T1WI, DWI, ADC, and magnetization transfer MRI [MTI]), were developed and validated in an independent test cohort. The RMs’ diagnostic performance was compared to that of visual interpretation using identical sequences and a clinical model.ResultsThe final population included 123 patients (81 men, 42 women; mean age: 30.26 ± 7.98 years; training cohort, n = 93; test cohort, n = 30). The area under the receiver operating characteristic curve (AUC) of RM1, RM2, and RM3 was 0.86 (p = 0.001), 0.88 (p = 0.001), and 0.93 (p = 0.02), respectively. The decision curve analysis confirmed a progressive improvement in the diagnostic performance of three RMs with the addition of more specific sequences. All RMs performance surpassed the visual interpretation based on the same MR sequences (visual model 1, AUC = 0.65, p = 0.56; visual model 2, AUC = 0.63, p = 0.04; visual model 3, AUC = 0.77, p = 0.002), as well as the clinical model composed of C-reactive protein and erythrocyte sedimentation rate (AUC = 0.60, p = 0.13).ConclusionsThe RMs, utilizing various combinations of conventional, DWI and MTI sequences, significantly enhance radiologists’ ability to accurately characterize intestinal fibrosis in patients with CD.Critical relevance statementThe utilization of MRI-based RMs significantly enhances the diagnostic accuracy of radiologists in characterizing intestinal fibrosis.Key PointsMRI-based RMs can characterize CD intestinal fibrosis using conventional, diffusion, and MTI sequences.The RMs achieved AUCs of 0.86–0.93 for assessing fibrosis grade.MRI-radiomics outperformed visual interpretation for grading CD intestinal fibrosis.Graphical

Machine learning prediction of prognosis in lung cancer with inter-institutional generalizability: A multicenter cohort study (WJOG15121L: REAL-WIND).

8571 Background: Recent advances in machine learning have improved the accuracy of prognosis prediction of cancer. The applicability of algorithm creation methods to data from various institutions that were not utilized for algorithm training should be established to create an ML algorithm suitable for clinical practice for patients at various facilities. However, no large-scale studies have validated the inter-institutional generalizability of this algorithm across different institutions. Methods: We conducted a multicenter, retrospective, hospital-based cohort study of consecutive patients diagnosed with stage IV lung cancer between January 2016 and December 2020. The study population was randomly categorized into the training and independent test cohorts with a 2:1 ratio. The primary metric to assess algorithm performance was the area under the receiver operating characteristic curve in the independent test cohorts. To assess the inter-institutional generalizability of the algorithm, we investigated its ability to predict patient outcomes in the remaining facility after being trained using data from 15 other facilities. Results: We enrolled 6751 patients with stage IV lung cancer from 16 institutions in Japan. The median age was 70 years, and 4,421 (65%) had a PS score of 0 or 1. As a driver oncogene, 1,515 (22%) patients had mutated EGFR. The algorithm performance metrics in the test cohort showed that the area under the receiver operating characteristic curve values ranging from 0.83 to 0.90 across different predicted survival days (180 to 1080). The performance test of 16 algorithms for investigating inter-institutional generalizability showed median areas under the receiver operating characteristic curve of 0·87 (range, 0·84–0·92), 0·84 (range, 0·78–0·88), 0·84 (range, 0·76–0·89), and 0·84 (range, 0·75–0·90) at 180, 360, 720, and 1080 days, respectively. Specifically, at 180 predicted survival days, the AUC for the performance of algorithms classified based on facilities exceeded 0.8 for all algorithms (16/16). Similarly, the AUC exceeded 0.8 for 15/16, 12/16, and 12/16 facilities at 360, 720, and 1080 days, respectively, indicating a high degree of inter-institutional generalizability of the algorithm creation method. Conclusions: This large-scale study showed that our ML algorithm can accurately predict prognosis in patients with stage IV lung cancer. Furthermore, the algorithm creation method demonstrated a high degree of inter-institutional generalizability. Therefore, our ML algorithm can be generalized to various hospitals and implemented in clinical practice to benefit patients across diverse healthcare facilities.

The use of micronuclei DNA from erythrocytes for the detection of early-stage gastrointestinal cancers.

4085 Background: Gastrointestinal (GI) cancers account for over a third of cancer deaths. However, their early detection remains a challenge. Micronuclei (MN) are extranuclear bodies containing damaged chromosome segments indicative of genomic instability. Increased MN frequency in hematopoietic cells is observed in patients with solid tumors. We have developed a method (WO2021/228246 A1) for purifying and characterizing micronuclei DNA (MN-DNA) in erythrocytes from peripheral blood. By comparing MN-DNA from healthy donors (HDs) and GI cancer patients, we identified significant changes in read densities at specific genomic locations in patients, which were termed as tumor-associated MN-DNA (taMN-DNA) signatures. Here, we evaluated the potential of these signatures for early-stage GI cancer detection, including colorectal (CC), gastric (GC) and esophageal (EC) cancers. Methods: Peripheral blood (1-2 ml) was collected from healthy donors (HD) and cancer patients MN-DNA isolation and purification from erythrocytes. Participants were randomly divided into training, validation and independent test cohorts in a 7:2:1 ratio, maintaining similar cancer types, CC/GC/EC/HD ratio, gender and age distribution. Distinct MN-DNA features between cancer patients and HDs were identified using sequencing data. Logistic regression algorithms were applied using these features for cancer detection. Results: The study enrolled 1753 participants, including 987 HDs, 420 CC, 282 GC and 64 EC cases. Of these, over half were diagnosed with early-stage diseases; TNM stage 0-I accounted for 40.57%, stage II for 23.32%, stage III for 29.25% and stage IV for 6.87%. The cancer detection model with MN-DNA features yielded an overall accuracy of 87.43%, with a sensitivity of 83.66% and specificity of 90.36%. For individual caner types, sensitivity was 85.37% for CC, 79.31% for GC and 92.31% for EC. For early-stage (stage 0-II) CC, GC and EC, the model demonstrated sensitivities of 95.12%, 86.05% and 100.00%, respectively, at 90.36% specificity. Conclusions: Unlike cell-free DNA, MN-DNA are chromosomal fragments in the cytoplasm. The abundance of erythrocytes in peripheral blood provides an easily accessible source for MN-DNA enrichment. This pilot study illustrates the potential of MN-DNA as a tool for early GI cancer detection, contributing to large-scale efforts towards developing an effective GI cancer screening test.

A multi-center study for colorectal cancer early detection among patients with high-risk disease using a cell-free fragmentomics assay.

3613 Background: Colorectal cancer (CRC) is the third most prevalent and the second deadliest cancer globally, accounting for an estimated 1,926,000 new cases and 903,800 deaths in the year 2022 alone. It's also established that chronic inflammatory conditions, such as Crohn's disease and ulcerative colitis, elevate the risk of developing CRC. Therefore, a timely CRC diagnosis for patients with colorectal disease can significantly improve their prognosis and therefore reduce cancer-related mortalities. Methods: In this study, we enrolled a total of 167 patients diagnosed with CRC (stages I: 43, II: 55, III: 56, IV: 13) and 217 patients with benign colorectal disease (BCD), from five participating hospitals. All participants underwent colonoscopy procedures, and their CRC or BCD status was pathologically confirmed. The development of the cancer detection model was based on a training cohort of 98 CRC patients and 131 BCD patients from three hospitals, and the evaluation of the model's performance was carried out on an independent test cohort comprising 69 CRC patients and 96 BCD patients from the remaining two hospitals. The predictive model leveraged cell-free fragmentomics profiling, employing low-pass whole-genome sequencing (WGS) on pre-operative plasma samples to identify CRC. Results: The early detection model exhibited remarkable efficacy in differentiating CRC patients from those with BCD, achieving an Area Under the Curve (AUC) of 0.930 in the training cohort and a nearly equivalent AUC of 0.926 in the independent test cohort. By applying the same cutoff determined in the training cohort, our model demonstrated a high sensitivity (91.8% in the training cohort and 91.3% in the test cohort) and a satisfactory specificity (78.6% in the training cohort and 82.3% in the test cohort). Notably, the model showed consistent and robust performance across various CRC characteristics, including tumor location, histology subtypes, grades, and mismatch repair deficiency status in both cohorts. Moreover, our model outperformed traditional CRC screening methods such as the fecal immunochemical test, carcinoembryonic antigen (CEA), and CA19-9, which achieved AUCs of 0.742, 0.720, and 0.543, respectively. Further validation was conducted using an additional cohort of 31 patients with advanced colorectal adenoma (advCRA), where the model achieved an AUC of 0.846 and a 66.7% sensitivity for distinguishing advCRA from BCD. Conclusions: The study introduces a non-invasive, cell-free fragmentomics assay that incorporates low-pass WGS and machine learning algorithms, demonstrating significant clinical potential in accurately distinguishing CRC and BCD patients. Our innovative approach offers a promising alternative to traditional screening methods that can improve early CRC detection, enhance CRC patient outcomes, and reduce mortality rates.