Imaging-molecular integration in veterinary oncology: a perspective on evidence standards, validation, and translational readiness

Abstract Veterinary oncology increasingly generates imaging, pathology, and molecular data, but these layers are often interpreted in parallel rather than linked within clinically testable models. Radiogenomics is used here as a focused form of imaging-molecular integration, whereas multimodal artificial intelligence refers more broadly to models combining imaging, pathology, molecular, biomarker, and clinical variables. Quantitative imaging, radiomics, and pathomics can describe tumour morphology and spatial heterogeneity, whereas molecular biomarkers may provide biological context relevant to prognosis and treatment response. The key challenge is not to add more artificial intelligence to veterinary cancer diagnostics, but to determine when integration adds value beyond clinical, imaging-only, and molecular-only baselines. This Perspective proposes minimum evidence standards for veterinary oncologic radiogenomics and multimodal artificial intelligence, covering animal- and lesion-level linkage, preprocessing, data-leakage control, justified multimodal fusion design, clinically meaningful comparators, calibration, external validation, cautious interpretation of explainability outputs, and transparent reporting. Imaging-molecular artificial intelligence should be treated as a translational research programme, not an immediately deployable diagnostic product.

Authors

Publication Details

Journal
Veterinary oncology.
Published
2026-11-15
DOI
https://doi.org/10.1186/s44356-026-00080-5
Primary Topic
Radiomics and Machine Learning in Medical Imaging
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Imaging-molecular integration in veterinary oncology: a perspective on evidence standards, validation, and translational readiness

Soroush Akbari‐Ardabili
Veterinary oncology.
Radiomics and Machine Learning in Medical Imaging
article

Imaging-molecular integration in veterinary oncology: a perspective on evidence standards, validation, and translational readiness

Soroush Akbari‐Ardabili
article en

Abstract

Abstract Veterinary oncology increasingly generates imaging, pathology, and molecular data, but these layers are often interpreted in parallel rather than linked within clinically testable models. Radiogenomics is used here as a focused form of imaging-molecular integration, whereas multimodal artificial intelligence refers more broadly to models combining imaging, pathology, molecular, biomarker, and clinical variables. Quantitative imaging, radiomics, and pathomics can describe tumour morphology and spatial heterogeneity, whereas molecular biomarkers may provide biological context relevant to prognosis and treatment response. The key challenge is not to add more artificial intelligence to veterinary cancer diagnostics, but to determine when integration adds value beyond clinical, imaging-only, and molecular-only baselines. This Perspective proposes minimum evidence standards for veterinary oncologic radiogenomics and multimodal artificial intelligence, covering animal- and lesion-level linkage, preprocessing, data-leakage control, justified multimodal fusion design, clinically meaningful comparators, calibration, external validation, cautious interpretation of explainability outputs, and transparent reporting. Imaging-molecular artificial intelligence should be treated as a translational research programme, not an immediately deployable diagnostic product.

Veterinary oncology.Vol. 3(1)
Radiomics and Machine Learning in Medical Imaging
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon