AI and Cancer Research ── 2026-11-13
New papers on cancer and AI, collected from journal feeds, PubMed (searched across all journals) and OpenAlex (a public index used instead of Google Scholar), judged by Jev (an AI that makes language judgements): original research, AI central, which of 13 themes, treatment or trials, and then ranked by importance up to a daily limit. Journals are not filtered by impact factor; each paper shows its journal's IF-equivalent (OpenAlex two-year mean citedness) and the date of that value. Each entry shows the full original abstract.
All cancers: Cancer biology(2)
Hematologic malignancies: AML(0) · CML(0) · ALL(0) · CLL(0) · Lymphoma(1) · Myeloma(0)
Solid tumors: Breast(1) · Lung(1) · Colorectal(1) · Pancreatic(1) · Prostate(0) · Brain(0) · Ovarian(0) · Endometrial(1) · Gastric(0) · Liver(2) · Kidney(1) · Bladder(1)
Rare cancers: ACC(0)
Cancer molecular biology and AI
- Abstract
Message Artificial intelligence (AI) has proven effective in assessing bowel preparation adequacy, but its role in guiding personalised laxative dosing is untested. In this prospective, multicentre, endoscopist-masked randomised trial, 1650 participants were assigned to AI-assisted or conventional groups using self-evaluation (SE) by patients during a 3 L split-dose polyethylene glycol (PEG) regimen. The AI group used significantly less PEG (mean difference –496.1 mL; p<0.001) while maintaining similar bowel cleanliness, polyp detection rates (PDR) and withdrawal times. Right colon scores were slightly higher, with less nausea and bloating reported in the AI group. AI-assisted assessment (AIA) offers an effective personalised approach to colonoscopy and warrants evaluation with other regimens. In more detail Colonoscopy plays a vital role in early detection of colorectal cancer (CRC), with adequate bowel cleansing being essential for mucosal visualisation.1 Our team previously developed an AIA tool capable of real-time evaluation of rectal effluent...
Journal IF-equivalent: 3.5 (OpenAlex 2-year mean citedness, value as of 2026-11-11, retrieved 2026-11-12; not the official Clarivate IF)Reference: [No authors listed]. Artificial intelligence-tailored bowel preparation reduces laxative burden and promotes sustainable endoscopy. Gut. [Epub ahead of print]. doi:.Checked: Abstract only - Abstract
PURPOSE: To develop and validate an artificial intelligence-enabled platform that combines efficient screening with explicit eligibility logic by converting unstructured cancer trial criteria into reusable structured queries, and illustrates potential eligibility using synthetic patients. METHODS: We downloaded actively recruiting US interventional treatment trials for advanced/metastatic breast cancer, colon cancer, and non-small cell lung cancer from ClinicalTrials.gov. Medical oncologists created 24 synthetic patient vignettes. A large language model converted trial eligibility criteria into Structured Query Language (SQL) code and patient information into structured records, enabling rule-based matching without further LLM inference. Cancer details and treatment history were considered, but not laboratory results or comorbidities. Validation included physician editing of generated eligibility code for 30 trials and independent assessments by two physicians, followed by consensus adjudication, for eligibility across ten trials. We descriptively summarized potential eligibility by trial characteristics and when varying age, ECOG performance status, sex, and ZIP code. RESULTS: Of 833 candidate trials, 746 met inclusion criteria. Performing matching for all 746 trials took a median of 0.459 seconds per patient, excluding patient-data preparation. Against physician consensus, automated matching agreed on 230/240 patient-trial pairs (95.8%; sensitivity 83.3%, specificity 100%). Across synthetic patients, potentially eligible trials ranged from 10 to 258 without geographic restrictions. Counts decreased with worse performance status and geographic restriction and were relatively stable across ages 30-90. Later-phase, randomized, and molecularly selective trials generally matched fewer vignettes. CONCLUSION: LLM-generated SQL enables rapid, explicit rule-based screening for potential trial eligibility. Evaluation in synthetic vignettes supports feasibility, while missed matches underscore the need for further validation.
Journal IF-equivalent: 4.4 (OpenAlex 2-year mean citedness, value as of 2026-11-11, retrieved 2026-11-12; not the official Clarivate IF)Reference: Goel K, Myall NJ, Dickerson J, Caswell-Jin JL, Johnson T, Worth JE, et al. Quantifying cancer clinical trial eligibility using artificial Intelligence-Based matching. JNCI Cancer Spectrum. 2026 Nov 10 [Epub ahead of print]. doi:10.1093/jncics/pkag120.Checked: Abstract only
See all 2 Cancer biology papers →
Hematologic malignancies
Acute myeloid leukemia (AML): treatment, trials and AI
No new qualifying paper for this issue.
Chronic myeloid leukemia (CML): treatment, trials and AI
No new qualifying paper for this issue.
Acute lymphoblastic leukemia (ALL): treatment, trials and AI
No new qualifying paper for this issue.
Chronic lymphocytic leukemia (CLL): treatment, trials and AI
No new qualifying paper for this issue.
Lymphoma: treatment, trials and AI
- Abstract
[18F] FDG PET/CT is central to lymphoma staging and treatment-response assessment, but repeated examinations increase cumulative radiation exposure, largely from the CT component. This retrospective study included 152 patients with 211 whole-body PET/CT scans and evaluated whether deep learning-denoised simulated low-dose CT can preserve diagnostic performance. A single-scan cohort of 108 patients was used exclusively for model development, while an independent follow-up cohort of 44 patients (103 examinations) was reserved for quantitative evaluation and a fully crossed multi-reader, multi-case study. Simulated low-dose CT images ranging from 2/3 to 1/8 of the standard dose were reviewed, and the 1/4-dose level was selected as the lowest clinically acceptable level. A denoising model was trained using paired standard-dose CT (SDCT) and simulated low-dose CT at 1/4 of the standard dose (SLDCT) images. Six nuclear medicine physicians evaluated three imaging conditions: PET/CT with SDCT, PET/CT with SLDCT, and PET/CT with denoised SLDCT (D-SLDCT), for image quality, lesion detectability, staging, and treatment response. D-SLDCT significantly improved quantitative image quality compared with SLDCT (all p 2 and among junior readers. Lesion detectability and treatment-response assessment did not differ significantly across conditions, while within-reader lesion-detection agreement with the SDCT condition improved after denoising (κ = 0.72 vs 0.67). The D-SLDCT condition preserved diagnostic performance, supporting radiation-conscious lymphoma PET/CT. This study was registered with the Chinese Clinical Trial Registry (ChiCTR) on February 14, 2026 (registration number: ChiCTR2600119037).
Journal IF-equivalent: 3.9 (OpenAlex 2-year mean citedness, value as of 2026-10-13, retrieved 2026-10-15; not the official Clarivate IF)Reference: He Y, Peng H, Li Z, Zou W, Xiang F, Li Y, et al. Feasibility of Simulated Low-Dose CT with Deep Learning Denoising for Lymphoma PET/CT Imaging: A Clinical Study on Lesion Detection, Staging, and Therapy Response. J Digit Imaging Inform med. 2026 Nov 9 [Epub ahead of print]. doi:10.1007/s10278-026-02474-1.Checked: Abstract only
Multiple myeloma: treatment, trials and AI
No new qualifying paper for this issue.
Solid tumors
Breast cancer: treatment, trials and AI
- Abstract
Lymphovascular invasion (LVI) is a key histopathological factor driving the metastatic process in invasive breast cancer; however, a definitive diagnosis is typically only possible postoperatively. Radiomics based on dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) has shown promise for the preoperative prediction of LVI. Nevertheless, existing studies are largely single-center, lack systematic comparisons of machine learning algorithms, and have insufficient external validation. This study developed and validated multicenter DCE-MRI radiomics models for preoperative LVI prediction, systematically compared ten machine learning algorithms, and quantified the incremental value of radiomics over clinical variables available before surgery. This retrospective multicenter study included 935 female patients with pathologically confirmed invasive ductal carcinoma from three institutions. All patients underwent preoperative 3.0 T DCE-MRI. After manual three-dimensional delineation of regions of interest on DCE-MRI images, 1,197 radiomic features were extracted with PyRadiomics after image resampling to 2 × 2 × 2 mm³ (B-spline interpolation), intensity normalization, and fixed-bin-width discretization. Feature selection (Mann-Whitney U test, Pearson correlation pruning with |r| > 0.9, and least absolute shrinkage and selection operator [LASSO] regression) was confined to the training cohort. Ten classifiers were compared: logistic regression (LR), support vector machines (SVM), k-nearest neighbors (KNN), random forests (RF), Extremely Randomized Trees (ExtraTrees), XGBoost, LightGBM, Gradient Boosting, AdaBoost, and multi-layer perceptron (MLP). A primary clinical model was built using only variables available before surgery (age, histological grade, Ki-67, MRI axillary lymph node status, and MRI tumor size); a sensitivity clinical model additionally included postoperative sentinel lymph node (SLN) status. Combined clinical-radiomics models and a nomogram were constructed, and all probability thresholds were fixed at the value maximizing the Youden index in the training set. Performance was assessed by the area under the receiver operating characteristic curve (AUC) with the DeLong test, calibration intercept and slope, Brier score, and decision curve analysis (DCA) in a training set, an internal validation set, and two independent external test sets. In the multivariable analysis, SLN status (odds ratio [OR] 5.743, 95% confidence interval [CI]: 3.693–8.931, p < 0.001) and MRI tumor size (OR 1.413, 95% CI: 1.186–1.684, p < 0.001) were independently associated with LVI; because SLN status is established intraoperatively or postoperatively, the primary clinical model used only the five preoperative variables. Among the ten radiomics algorithms, LR and MLP showed the most stable cross-center performance (LR: AUC 0.841, 0.716, 0.632, and 0.586 in the training, internal validation, test 1, and test 2 cohorts, respectively), whereas complex ensemble methods overfit (Gradient Boosting: 0.991 in training vs. 0.696 and 0.532 externally; XGBoost: 0.894 vs. 0.613 and 0.453). The preoperative combined model achieved AUCs of 0.856, 0.721, 0.665, and 0.603, respectively, significantly outperforming the preoperative clinical model in the training, validation, and test 2 cohorts (DeLong p < 0.001, p = 0.013, and p < 0.001) but not in test 1 ( p = 0.306), and significantly outperforming the radiomics model only in test 1 ( p = 0.026). The incremental value of radiomics was significant in the training cohort (integrated discrimination improvement p < 0.001) but not in the external cohorts. Calibration slopes of the combined model declined externally (0.49, 0.28, and 0.21), indicating that probability estimates require local recalibration. In this multicenter study, LR and MLP were the most robust radiomics classifiers for preoperative LVI prediction, whereas complex ensemble algorithms were prone to overfitting on external validation. The preoperative combined model improved discrimination and clinical net benefit over the clinical model internally, but its external incremental value was inconsistent and its calibration deteriorated in the external cohorts. A sensitivity analysis using postoperative SLN status showed higher absolute performance, underscoring that genuinely preoperative tools trade performance for applicability. These findings emphasize rigorous multicenter benchmarking, recalibration before local deployment, and honest reporting of external performance in radiomics research.
Journal IF-equivalent: 5.9 (OpenAlex 2-year mean citedness, value as of 2026-11-04, retrieved 2026-11-05; not the official Clarivate IF)Reference: Li H, Zeng Y, Xiao Q, Wang W, Zhang Y. Preoperative prediction of lymphovascular invasion in breast cancer using multicenter DCE-MRI radiomics and clinical features. Sci Rep. 2026 Nov 10 [Epub ahead of print]. doi:10.1038/s41598-026-75438-8.Checked: Abstract only
Lung cancer: treatment, trials and AI
- Abstract
BACKGROUND: Accurate characterization of pulmonary subsolid nodules (SSNs) is important for selecting surveillance, biopsy, or surgery. Existing clinical models were developed in broader nodule populations and may not fully reflect the radiological and biological features of SSNs. METHODS: This dual-center retrospective study included 225 patients with pathologically confirmed SSNs from January 2018 to January 2024. The development center contributed 167 patients, split into internal training (n = 118) and internal validation (n = 49) cohorts, and the external center contributed 58 patients. Ten machine-learning algorithms were compared within the development cohort. The final combined model used four routinely available predictors: spiculation, consolidation-to-tumor ratio (CTR), CYFRA 21 - 1, and nodule diameter. Discrimination, calibration, decision-curve analysis, pairwise model comparison, and sensitivity analyses were performed. RESULTS: The combined model achieved AUCs of 0.932 (95% CI, 0.887-0.977), 0.954 (95% CI, 0.895-1.000), and 0.930 (95% CI, 0.864-0.995) in the internal training, internal validation, and external validation cohorts, respectively. In external validation, the corresponding AUCs for the Mayo, VA, Brock, and PKUPH models were 0.750, 0.560, 0.800, and 0.840. Calibration was evaluated in all three cohorts, and decision-curve analysis showed net benefit across threshold probabilities of approximately 0.12-0.68 in external validation. CONCLUSIONS: A four-variable combined model integrating imaging features and CYFRA 21 - 1 showed consistent discrimination across the internal and external validation cohorts. The web calculator may support research-stage risk estimation, but prospective multicenter validation is needed before clinical implementation.
Journal IF-equivalent: 3.4 (OpenAlex 2-year mean citedness, value as of 2026-11-11, retrieved 2026-11-12; not the official Clarivate IF)Reference: Fu Y, Tu H. Development and validation of a machine learning-based combined model for malignancy risk prediction in pulmonary subsolid nodules: a dual-center study. BMC Pulm Med. 2026 Aug 13;26(1):544. doi:10.1186/s12890-026-04594-4.Checked: Abstract only
Colorectal cancer: screening, treatment, trials and AI
- Abstract
BACKGROUND: In clinical practice, approximately 15-25% of patients with locally advanced rectal cancer (LARC) achieve a pathological complete response (pCR), defined as tumor regression grade 0 according to the AJCC 8th edition, which is closely associated with favorable prognosis. However, noninvasive and accurate preoperative assessment methods remain a major clinical challenge. We aimed to develop an automated artificial intelligence (AI) system for predicting the response to neoadjuvant chemoradiotherapy and prognosis in LARC using pre-treatment T2WI. METHODS: This multicenter retrospective study included 227 LARC patients from Center 1 as the development cohort and 72 and 55 patients from two separate centers as external validation cohorts 1 and 2, respectively. 3D Swin-UNETR was constructed to automatically identify LARC tumors. Based on the segmented tumor volumes, three analytical approaches were implemented to support downstream tasks. Specifically, 1,132 radiomic features (Rf) were extracted using PyRadiomics, while 2,048 deep learning features (Dlf) were derived by 3D ResNet-152 architecture. Following a series of feature selection steps, predictive response scores were generated, including RF scores (Rfs) and Dlf scores (Dlfs). Additionally, an end-to-end analysis was performed to automatically generate Dl scores (Dls). A multi-perspective feature engineering model was then developed by integrating Rfs, Dlfs, and Dls. The predictive performance was evaluated using the area under the receiver operating characteristic curves (AUROCs). Prognostic indices, including disease-free survival (DFS) and overall survival (OS), were assessed using Kaplan-Meier (KM) analysis, with the Log-Rank test used to compare KM curves. RESULTS: The fusion model achieved moderate AUROCs of 0.756 and 0.751 in external validation cohorts 1 and 2, respectively. It effectively stratified OS risk, with p-values < 0.05 across all cohorts. For DFS, statistical significance was inconsistent across cohorts, being achieved in the development cohort (p = 0.0084) and external cohort 2 (p = 0.044), but not in external cohort 1 (p = 0.068). These findings are exploratory and require prospective confirmation. CONCLUSIONS: The proposed AI system demonstrates moderate, externally validated performance for assessing treatment response and OS risk stratification in LARC and may serve as a decision-support tool to assist clinical judgment.
Journal IF-equivalent: 3.8 (OpenAlex 2-year mean citedness, value as of 2026-11-03, retrieved 2026-11-05; not the official Clarivate IF)Reference: Sun K, Lu S, Li X, Wang Y, Jiang Z, Wang X, et al. Pre-treatment T2WI-based automated artificial intelligence system for predicting neoadjuvant chemoradiotherapy response and prognosis in locally advanced rectal cancer: a multicenter study. BMC Med Imaging. 2026 Aug 11;26(1):601. doi:10.1186/s12880-026-02649-w.Checked: Abstract only
Pancreatic cancer: treatment, trials and AI
- Abstract
BACKGROUND: Microscopic margin involvement (≤ 1 mm) strongly predicts poor prognosis in pancreatic ductal adenocarcinoma (PDAC). Adjuvant chemotherapy (ACT) after upfront resection fails to fully mitigate this risk. It remains uncertain whether neoadjuvant chemotherapy (NACT) can neutralize the adverse prognostic impact of impending ≤ 1 mm margins. METHODS: Prognostic significance of ≤ 1 mm margins was assessed in an upfront-resection cohort (n = 228) using multivariable Cox regression. A separate cohort (n = 110; 52 NACT, 58 upfront surgery) compared overall survival (OS) by margin status. A CT-based radiomics model using a Laplacian-Dirichlet operator was developed to predict ≤ 1 mm margins and externally validated (n = 230). RESULTS: ≤ 1 mm margins independently predicted worse OS vs. > 1 mm (15.0 vs. 24.0 months, p = 0.005). ACT numerically extended OS for ≤ 1 mm patients (16.0 vs. 15.0 months, p = 0.271) but did not close the survival gap. Notably, ≤ 1 mm patients receiving NACT achieved median OS of 29.4 months-comparable to > 1 mm upfront patients (31.6 months)-and significantly better than ≤ 1 mm upfront patients (19.6 months, p = 0.045). The model predicted ≤ 1 mm margins with AUCs of 0.772 (95% CI: 0.705-0.838) and 0.723 (95% CI: 0.657-0.789) in internal and external validation. CONCLUSION: NACT-not ACT-selectively rescues high-risk patients destined for ≤ 1 mm margins. Our radiomics model enables preoperative identification, supporting personalized neoadjuvant therapy.
Journal IF-equivalent: 2.4 (OpenAlex 2-year mean citedness, value as of 2026-11-11, retrieved 2026-11-12; not the official Clarivate IF)Reference: Li M, Lin Y, Wang Z, Liu J, Cai H, Liu X, et al. CT‐Based Radiomics for Preoperative Prediction of Surgical Margins in Pancreatic Head Cancer: A Tool for Selecting Patients for Neoadjuvant Chemotherapy. Journal of Surgical Oncology. 2026 Nov 9 [Epub ahead of print]. doi:10.1002/jso.70453.Checked: Abstract only
Prostate cancer: treatment, trials and AI
No new qualifying paper for this issue.
Brain tumours: treatment, trials and AI
No new qualifying paper for this issue.
Ovarian cancer: treatment, trials and AI
No new qualifying paper for this issue.
Endometrial cancer: treatment, trials and AI
- Abstract
Preoperative identification of p53-abnormal (p53abn) endometrial cancer is important for molecular risk stratification, yet biopsy-based testing and p53 immunohistochemistry may be limited by sampling and interpretive variability. This study aimed to develop and externally validate a dynamic contrast-enhanced MRI (DCE-MRI) Fusion model integrating peritumoral radiomics, intratumoral habitat features, and clinical predictors for noninvasive prediction of p53abn endometrial cancer. This retrospective dual-center study included 498 patients with surgically confirmed endometrial cancer and sequencing-based molecular classification. Patients from Institution 1 were assigned to a training cohort ( n = 351) and an internal validation cohort ( n = 86), and patients from Institution 2 formed an external test cohort ( n = 61). Radiomic features were extracted from the whole tumor, 3-mm and 5-mm peritumoral regions, and three unsupervised intratumoral habitat subregions. After reproducibility assessment, statistical filtering, correlation analysis, and least absolute shrinkage and selection operator selection, single-domain models were developed. A Fusion model combined clinical predictors with the 5-mm peritumoral and habitat radiomics scores. Performance was assessed using discrimination, calibration, decision curve analysis, reclassification metrics, and Shapley additive explanations (SHAP) interpretation. The Habitat model showed the most stable single-domain performance, with AUCs of 0.831, 0.813, and 0.788 in the training, internal validation, and external test cohorts, respectively. The Fusion model achieved AUCs of 0.891, 0.885, and 0.863, with sensitivities of 0.882, 0.765, and 0.889 and negative predictive values of 0.974, 0.925, and 0.968, respectively. Brier scores were 0.078, 0.095, and 0.110. SHAP analysis identified the Habitat and 5-mm peritumoral radiomics scores as the dominant contributors to prediction. A DCE-MRI Fusion model integrating clinical, peritumoral, and habitat radiomics demonstrated good discrimination for preoperative p53abn prediction. The model may serve as a noninvasive adjunct to molecular testing for preoperative risk stratification.
Journal IF-equivalent: 3.8 (OpenAlex 2-year mean citedness, value as of 2026-11-03, retrieved 2026-11-05; not the official Clarivate IF)Reference: Dai H, Guo W, Liu Y, Yan B, Lyu B, Wang F, et al. DCE-MRI-derived peritumoral and habitat radiomics for preoperative prediction of p53-abnormal endometrial cancer: a dual-center retrospective study with external test. BMC Med Imaging. 2026 Nov 10 [Epub ahead of print]. doi:10.1186/s12880-026-03045-0.Checked: Abstract only
See all 1 Endometrial papers →
Gastric cancer: treatment, trials and AI
No new qualifying paper for this issue.
Liver cancer: treatment, trials and AI
- Abstract
WHAT IS ALREADY KNOWN ABOUT THIS TOPIC The tumour–stroma ratio (TSR) is a promising prognostic factor in solid tumours, yet its application is limited by quantification challenges and conflicting prognostic reports. Current assessments typically rely on manual estimation or supervised segmentation artificial intelligence (AI), which treats TSR as a static scalar and assumes a linear relationship with prognosis. WHAT THIS STUDY ADDS Non-linear discovery: This study identified a distinct inverted U-shaped relationship between TSR and hepatocellular carcinoma (HCC) prognosis, defining a specific risk window that challenges the traditional linear assumption and explaining previous contradictory reports. Methodological innovation: We developed a foundation model (Conch v1.5)-driven framework that enables precise, annotation-free TSR quantification, showing strong correlation with expert pathologists’ delineation. Architectural innovation: We developed a novel ‘Token-Guided Multimodal Framework’ that transforms TSR from a simple scalar into a high-dimensional ‘semantic anchor’ alongside whole-slide image...
Journal IF-equivalent: 3.5 (OpenAlex 2-year mean citedness, value as of 2026-11-11, retrieved 2026-11-12; not the official Clarivate IF)Reference: [No authors listed]. Token-guided multimodal prognosis in hepatocellular carcinoma: a framework steered by tumour-stroma ratio. Gut. [Epub ahead of print]. doi:.Checked: Abstract only - Abstract
The aim of this study was to develop and validate a machine learning model based on contrast-enhanced CT (CECT) for predicting the overall survival (OS) in unresectable HCC patients treated with TACE combined with immune checkpoint inhibitors (ICIs) plus molecular targeted therapies (MTTs). This study retrospectively reviewed 180 unresectable HCC patients who received TACE-ICIs-MTTs therapy from 4 medical centers. The study comprised of training cohort ( n = 93), internal validation cohort ( n = 39) and external validation cohort ( n = 48). A series of analyses and kinds of models were conducted to build a Radscore. Then, models were constructed using Radscore and clinical variables selected through multivariate Cox regression analysis, evaluated by the Harrell’s concordance index (C-index) and the area under the curve (AUC). The radiomics-clinical model was further assessed by decision curve and calibration curve analysis. The Radscore was derived from 6 features selected by the optimal model (Lasso Cox + RSF with the highest C-index of 0.777). HBV infection and vascular invasion are clinical independent predictors of OS. The radiomics-clinical model demonstrated favorable performance. It achieved C-index values of 0.893(0.859–0.927), 0.758 (0.657–0.86) and 0.74 (0.624–0.856) in the training, internal validation, and external validation cohorts, respectively. In the external validation cohort, the AUCs achieved 0.731 (0.569–0.893), 0.7 (0.515–0.885), and 0.756 (0.569–0.943) for predicting 1-, 2-, and 3-year outcomes, respectively. The machine learning model based on CECT exhibited favorable performance in predicting overall survival in unresectable HCC patients treated with TACE-ICIs-MTTs, showing potential to identify likely responders and guide treatment decision-making.
Journal IF-equivalent: 5.9 (OpenAlex 2-year mean citedness, value as of 2026-11-04, retrieved 2026-11-05; not the official Clarivate IF)Reference: Ning S, Chen Y, Feng B, Sun Z, Qu Q, Lv Y, et al. Machine learning model using contrast-enhanced CT for predicting overall survival in unresectable hepatocellular carcinoma treated with TACE plus targeted and immunotherapy: a multicenter study. Sci Rep. 2026 Nov 10;16(1):37437. doi:10.1038/s41598-026-64608-3.Checked: Abstract only
Kidney cancer: treatment, trials and AI
- Abstract
This study aimed to develop and evaluate a multimodal predictive system (MPS) for assessment of renal sinus fat invasion (RSFI) in clear cell renal cell carcinoma (ccRCC) patients. A total of 875 ccRCC patients from four independent institutions were retrospectively analyzed. The cohort was divided into a training set ( n = 445), validation set ( n = 111), and two multicentre evaluation sets ( n = 181, n = 138). Radiomics features were extracted from the intratumoral ROI and tumor-plus-peritumoral ROIs expanded by 1, 3, and 5 mm. Multiple machine learning classifiers were constructed and evaluated. In parallel, 2D, 2·5D, and 3D deep learning (DL) models based on ResNet34 were trained. A clinical model was developed using significant variables. The representative models from each modality were integrated via stacking ensemble learning to construct the MPS. Diagnostic performance was evaluated by AUC, sensitivity, specificity, and accuracy, while model interpretability was analyzed using SHAP. The peri-3 mm ExtraTrees radiomics model achieved the highest AUC values of 0.753 (95% CI: 0.6741 - 0.8322) on test1 set and 0.751 (95% CI: 0.6610 - 0.8406) on test2 set. Among DL models, the 2·5D-Peri-5 mm model showed the most consistent numerical performance across the two multicentre evaluation cohorts with AUC values of 0.762 (95% CI: 0.6886 - 0.8360) on test1 set and 0.774 (95% CI: 0.6828 - 0.8651) on test2 set. The clinical model reached AUCs of 0.735 (0.6571 - 0.8121) and 0.736 (0.6499 - 0.8216) in the test sets. The MPS integrating the selected radiomics, DL, and clinical models via an SVM stacking strategy achieved the highest numerical AUCs, reaching 0.791 (95% CI: 0.7213–0.8607) in test1 and 0.777 (95% CI: 0.6894–0.8655) in test2. SHAP analysis revealed that features derived from the 2·5D-Peri-5 mm model contributed most significantly to RSFI prediction. The MPS may provide adjunctive quantitative information for preoperative RSFI risk stratification in patients with CT findings suspicious for RSFI, but prospective evaluation against existing expert assessment is required before clinical use.
Journal IF-equivalent: 5.7 (OpenAlex 2-year mean citedness, value as of 2026-11-04, retrieved 2026-11-05; not the official Clarivate IF)Reference: Lv D, Zhou H, Wang Q, Wang Z, Li J, Yao X, et al. An explainable AI system for preoperative prediction of renal sinus fat invasion in clear cell renal cell carcinoma: a multicenter study. Cancer Imaging. 2026 Nov 10 [Epub ahead of print]. doi:10.1186/s40644-026-01160-3.Checked: Abstract only
Bladder cancer: treatment, trials and AI
- Abstract
OBJECTIVES: This study aimed to develop and validate a deep learning model that integrates T2-weighted imaging (T2WI) and diffusion-weighted imaging (DWI) for the noninvasive prediction of muscle invasion in bladder cancer. MATERIALS AND METHODS: A total of 308 bladder tumors from 254 patients were retrospectively collected from three institutions. Of these, 177 tumors from Center 1 were partitioned via stratified sampling into a training set (n = 128) and an internal validation set (n = 49). The remaining two centers contributed 59 and 72 tumors, respectively, which served as external validation cohorts. A dual-branch 3D deep convolutional neural network (Joint model) based on biparametric MRI (bp-MRI) was developed using the training data and compared with a deep learning model based on T2WI (T2WI model) and VI-RADS assessments by radiologists. Diagnostic performance was evaluated using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity. RESULTS: The Joint model achieved mean AUCs of 0.979 ± 0.007 in the internal validation cohort, and 0.907 ± 0.013 and 0.916 ± 0.021 in the external validation cohorts. It outperformed the T2WI model in all cohorts and achieved higher AUCs than radiologists across all validation cohorts. Notably, for equivocal VI-RADS 2-3 lesions, the Joint model yielded higher accuracy than the T2WI model in external validation. CONCLUSION: Compared with the T2WI model and VI-RADS assessment, the Joint model demonstrated better diagnostic performance and generalizability for predicting muscle invasion in patients with bladder cancer. KEY POINTS: Question Reliable noninvasive tools for preoperative prediction of muscle invasion in bladder cancer are lacking, as TURBT-based staging carries procedural risks and frequent understaging.
Findings: A biparametric MRI deep learning model achieved AUCs of 0.907-0.979 across internal and external validation cohorts, outperforming T2WI-only models and VI-RADS. Critical relevance statement Contrast-free biparametric MRI combined with deep learning enables accurate, generalizable preoperative prediction of muscle invasion, offering particular value for equivocal VI-RADS 2-3 lesions and supporting noninvasive clinical decision-making.
Journal IF-equivalent: 6.5 (OpenAlex 2-year mean citedness, value as of 2026-11-03, retrieved 2026-11-05; not the official Clarivate IF)Reference: Liu Z, Hu Y, Lyu J, Deng L, Yang S, Huang X, et al. Biparametric MRI with deep learning for the preoperative prediction of muscle invasion in patients with bladder cancer: a multicenter study. Insights Imaging. 2026 Nov 9;17(1):289. doi:10.1186/s13244-026-02430-1.Checked: Abstract only
Rare cancers
Adenoid cystic carcinoma: treatment, trials and AI
No new qualifying paper for this issue.