Biology-Aware Harmonization Improves Machine Learning Models for Early Lung Cancer Detection in CT Scans
Researchers developed an improved CT radiomics-based machine learning approach for predicting lung cancer in pulmonary nodules that fall below the sensitivity threshold of standard clinical diagnosis. By augmenting training data with later-stage nodules and applying biology-aware harmonization to correct for imaging acquisition variability, models achieved ROC-AUC scores of 0.71–0.74, significantly outperforming near-chance baseline classifiers. The findings suggest that accounting for biological differences when combining datasets is critical for building effective early-detection AI tools in oncology.
A proof-of-principle study published on arXiv examined CT radiomics-based machine learning models designed to detect lung cancer in indeterminate pulmonary nodules at an early stage, before standard-of-care methods can reliably identify them. The core challenge was twofold: low malignancy rates in early-development nodules and variability introduced by different CT acquisition protocols. When classifiers were trained solely on ComBat-harmonized radiomic features from early-development nodules (n=106), performance was near chance. Augmenting the training set with later-development benign and malignant nodules (n=225) improved results only when harmonization accounted for biological differences across datasets—either by including a covariate distinguishing dataset types or by harmonizing each dataset separately. Biology-unaware harmonization with augmented data failed to produce consistent improvements. The best-performing approaches reached ROC-AUC values of 0.74 (95% CI: 0.69–0.79) and 0.71 (95% CI: 0.66–0.77), both statistically superior to baseline on both ROC-AUC and PR-AUC metrics. The authors caution that this is a small, single-center study and characterize it as a methodological proof-of-principle rather than a clinically validated tool.
What's missing
As a single-center, small-sample proof-of-principle study, key limitations include: lack of external validation on multi-center or prospective datasets; uncertainty about whether findings generalize across different CT scanner manufacturers or clinical settings; no comparison against existing clinical risk models (e.g., Lung-RADS, Mayo Clinic model); and the study does not address downstream clinical outcomes such as whether earlier detection translates to survival benefit.
What different sources said
- arXiv physicsCenter
Training Set Augmentation and Biology-Aware Harmonization Improve Radiomic Models for Lung Cancer Prediction in Indeterminate Nodules
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.