AI in Oncology
Research and news where artificial intelligence is the method — diagnosis, drug discovery, pathology, clinical decision support.
Multiparametric MRI-Based Ensemble Deep Learning for Preoperative Classification of Pituitary Neuroendocrine Tumor Subtypes.
Ndimulunde EN, et al
A multiparametric MRI-based ensemble deep learning model successfully classified pituitary neuroendocrine tumor (PitNET) subtypes noninvasively before surgery, achieving a test-set accuracy of 90.68% (95% CI: 84.7%–95.8%) and a macro-AUC of 0.966 (95% CI: 0.933–0.992) across four transcription facto…
Source →Artificial Intelligence-Estimated Intrapancreatic Fat and Its Relationship to Familial and Germline Pancreatic Cancer Risk.
Konikoff T, et al
A deep-learning AI pipeline for whole-pancreas MRI analysis found no significant difference in intrapancreatic fat (IPF) between 85 high-risk individuals (HRIs) with a family history of pancreatic cancer or pathogenic germline variants and 170 age-, sex-, BMI-, and diabetes-matched controls (mean pr…
Source →Deep learning auto-contouring of target and organs-at-risk on post-catheter implant CT images for prostate HDR brachytherapy.
Wallat EM, et al
A custom deep learning model based on the nnU-Net architecture outperformed a commercially available contouring solution for prostate high-dose-rate (HDR) brachytherapy, achieving a Dice similarity coefficient of 0.91 for the prostate planning target volume compared to 0.78 for the commercial model,…
Source →Mapping structural aging across human tissues reveals tissue-specific trajectories and coordinated deterioration.
Yadav A, et al
A new computational framework called PathStAR, applied to 25,306 post-mortem biopsies from 40 tissues in 970 donors aged 21–70 years, has revealed that human organs age structurally along distinct, nonlinear trajectories rather than deteriorating uniformly over time. Vascular tissue structural aging…
Source →Artificial intelligence-assisted design of MAGE-A4 ×CD16 T-cell receptor-like natural killer cell engagers.
Liu Y, et al
Using artificial intelligence-assisted computational structural prioritization combined with yeast-display screening, researchers engineered a novel natural killer cell engager (NKCE) targeting MAGE-A4, a cancer-testis antigen highly expressed in multiple malignancies but restricted in normal tissue…
Source →Exploratory immunomonitoring during radiochemotherapy in HNSCC and machine-learning reveal immune parameters associated with disease-free survival.
Donaubauer A, et al
A machine-learning analysis of 45 peripheral blood immune parameters in 70 head and neck squamous cell carcinoma (HNSCC) patients undergoing postoperative radio(chemo)therapy identified a 29-parameter immune signature capable of predicting disease-free survival, with the best model achieving a Matth…
Source →Comparison and interpretability of machine learning algorithms to predict survival of patients with prostate cancer.
Kim IE, et al
A machine learning study using the SEER 17 Database found that XGBoost achieved the best area under the curve, concordance index, and Brier score for predicting both overall survival and prostate cancer-specific survival among all models tested, outperforming traditional Cox proportional hazards mod…
Source →CT radiomics-histopathological correlation for mediastinal lymph node staging in non-small cell lung cancer.
Legrand J, et al
A machine learning radiomics model built on contrast-enhanced CT scans can predict malignancy in mediastinal lymph nodes of non-small cell lung cancer (NSCLC) patients with clinically positive nodal disease (cN+), achieving an area under the curve (AUC) of 0.703 using a random forest algorithm and 0…
Source →Application of Deep Learning Algorithm-Based Image Reconstruction in Improving Abdominal CT Image Quality in Adrenal and Renal Lesions Detection.
Tian J, et al
A deep learning-based CT image reconstruction algorithm called Precise Image (PI) significantly outperformed traditional filtered back projection (FBP) and iterative reconstruction (iDose) methods in abdominal CT image quality across all evaluated metrics in a retrospective study of 191 patients. No…
Source →Single-cell and spatial transcriptomic profiling identifies a fibroblast-like endothelial state and the MDK-NCL signaling axis in prostate cancer microenvironment remodeling.
Wang R, et al
Researchers have identified a fibroblast-like endothelial (FLE) cell population within the prostate cancer tumor microenvironment and established the MDK-NCL signaling axis as a key driver of intercellular communication and adverse clinical outcomes. Using integrated single-cell RNA sequencing, spat…
Source →This page is educational — it is not medical advice and does not replace consultation with an oncologist. Diagnostic and treatment decisions are made solely by specialist physicians.