PapersFlow Research Brief

Health Sciences · Medicine

Radiomics and Machine Learning in Medical Imaging
Research Guide

What is Radiomics and Machine Learning in Medical Imaging?

Radiomics and Machine Learning in Medical Imaging is the quantitative analysis of medical images to extract advanced features, combined with machine learning for predictive modeling, particularly in cancer imaging and precision medicine.

This field encompasses 130,046 works focused on radiomics feature extraction from medical images and machine learning applications for tumor heterogeneity assessment. Key methods include texture analysis and high-throughput quantitative feature extraction from datasets. Growth data over the past five years is not available in the provided records.

Topic Hierarchy

100%
graph TD D["Health Sciences"] F["Medicine"] S["Radiology, Nuclear Medicine and Imaging"] T["Radiomics and Machine Learning in Medical Imaging"] D --> F F --> S S --> T style T fill:#DC5238,stroke:#c4452e,stroke-width:2px
Scroll to zoom • Drag to pan
130.0K
Papers
N/A
5yr Growth
890.8K
Total Citations

Research Sub-Topics

Why It Matters

Radiomics and machine learning enable precise tumor characterization in cancer imaging, supporting precision medicine applications. For instance, low-dose CT screening reduced lung-cancer mortality, as shown in "Reduced Lung-Cancer Mortality with Low-Dose Computed Tomographic Screening" (2011) with results from the National Lung Screening Trial (ClinicalTrials.gov NCT00047385). In glioma diagnosis, machine learning-driven radiomics on ¹⁸F-FDG PET provides systematic review and meta-analysis for improved classification, as detailed in recent preprints. Tools like pyradiomics facilitate reproducible feature extraction from 2D and 3D images, aiding clinical translation in radiation therapy and pathology, referenced in news on AI integration for prognosis and treatment response prediction.

Reading Guide

Where to Start

"Radiomics: Images Are More than Pictures, They Are Data" (Gillies et al., 2015) first, as it provides a foundational explanation of radiomics as quantitative data extraction from medical images, essential before machine learning applications.

Key Papers Explained

"Radiomics: Images Are More than Pictures, They Are Data" (Gillies et al., 2015) establishes quantitative feature extraction, built upon by "A survey on deep learning in medical image analysis" (Litjens et al., 2017) which reviews deep learning for image analysis including segmentation. "3D Slicer as an image computing platform for the Quantitative Imaging Network" (Fedorov et al., 2012) supports these with tools for quantitative imaging, while "UNet++: A Nested U-Net Architecture for Medical Image Segmentation" (Zhou et al., 2018) advances segmentation critical for radiomics pipelines. "High-performance medicine: the convergence of human and artificial intelligence" (Topol, 2018) connects to broader AI applications.

Paper Timeline

100%
graph LR P0["TNM Classification of Malignant ...
1987 · 16.1K cites"] P1["New response evaluation criteria...
2008 · 28.3K cites"] P2["The American Joint Committee on ...
2010 · 9.1K cites"] P3["Reduced Lung-Cancer Mortality wi...
2011 · 10.6K cites"] P4["3D Slicer as an image computing ...
2012 · 8.3K cites"] P5["A survey on deep learning in med...
2017 · 13.1K cites"] P6["UNet++: A Nested U-Net Architect...
2018 · 8.2K cites"] P0 --> P1 P1 --> P2 P2 --> P3 P3 --> P4 P4 --> P5 P5 --> P6 style P1 fill:#DC5238,stroke:#c4452e,stroke-width:2px
Scroll to zoom • Drag to pan

Most-cited paper highlighted in red. Papers ordered chronologically.

Advanced Directions

Recent preprints focus on reproducibility with "Radiomics Quality Score 2.0: towards radiomics readiness levels and clinical translation for personalized medicine" and "Open-radiomics: a collection of standardized datasets and a technical protocol for reproducible radiomics machine learning pipelines." Guidelines in "Robust radiomics: a review of guidelines for radiomics in medical imaging" address robust pipelines. News highlights machine learning-based MRI radiomics for IL18 prediction in gliomas and FDA approvals for computational pathology.

Papers at a Glance

# Paper Year Venue Citations Open Access
1 New response evaluation criteria in solid tumours: Revised REC... 2008 European Journal of Ca... 28.3K
2 TNM Classification of Malignant Tumours 1987 16.1K
3 A survey on deep learning in medical image analysis 2017 Medical Image Analysis 13.1K
4 Reduced Lung-Cancer Mortality with Low-Dose Computed Tomograph... 2011 New England Journal of... 10.6K
5 The American Joint Committee on Cancer: the 7th Edition of the... 2010 Annals of Surgical Onc... 9.1K
6 3D Slicer as an image computing platform for the Quantitative ... 2012 Magnetic Resonance Ima... 8.3K
7 UNet++: A Nested U-Net Architecture for Medical Image Segmenta... 2018 Lecture notes in compu... 8.2K
8 Radiomics: Images Are More than Pictures, They Are Data 2015 Radiology 7.8K
9 QuPath: Open source software for digital pathology image analysis 2017 Scientific Reports 7.8K
10 High-performance medicine: the convergence of human and artifi... 2018 Nature Medicine 7.0K

In the News

Code & Tools

Recent Preprints

Machine Learning–Driven Radiomics for Glioma Diagnosis

linkedin.com Preprint

Our new paper, “Machine Learning–Driven Radiomics on ¹⁸F-FDG PET for Glioma Diagnosis: A Systematic Review and Meta-Analysis,” is now published in Cancer Imaging.

Radiomics Quality Score 2.0: towards radiomics readiness levels and clinical translation for personalized medicine

Sep 2025 nature.com Preprint

Radiomics is a tool for medical imaging analysis that could have a relevant role in precision oncology by offering precise quantitative support for clinical decision-making. The Radiomics Quality S...

Robust radiomics: a review of guidelines for radiomics in medical imaging

Jan 2026 frontiersin.org Preprint

**Introduction:** Radiomics aims to develop image-based biomarkers by combining quantitative analysis of medical images with artificial intelligence (AI) through a robust, reproducible pipeline. Sc...

Open-radiomics: a collection of standardized datasets and a technical protocol for reproducible radiomics machine learning pipelines

Aug 2025 bmcmedimaging.biomedcentral.com Preprint

As an important branch of machine learning pipelines in medical imaging, radiomics faces two major challenges namely reproducibility and accessibility. In this work, we introduce open-radiomics, a ...

Machine Learning–Driven radiomics on 18 F-FDG PET for glioma diagnosis: a systematic review and meta-analysis

Aug 2025 cancerimagingjournal.biomedcentral.com Preprint

Radiomics has emerged as a transformative approach to overcome these challenges by extracting high-dimensional, quantitative features from medical images that characterize tumor intensity, texture,...

Latest Developments

Recent developments in radiomics and machine learning in medical imaging research as of February 2026 include a conceptual shift towards using imaging for prognostic and predictive purposes through quantitative pattern extraction, with advancements highlighted in 2026 (Open Medscience). Additionally, there is ongoing research integrating multimodal deep learning radiomics for traditional Chinese medicine classification, and new AI-driven tools are expected to become standard in imaging departments to streamline workflows (GeneOnline, Beekley Blog). Notably, the Radiomics Quality Score 2.0 has been introduced to improve clinical translation, and there are active efforts to standardize datasets and pipelines for reproducible radiomics machine learning (Nature, BMC Medical Imaging). Furthermore, specialized conferences and courses are focusing on AI applications, deep learning, foundation models, and synthetic data in radiomics (Imaging.org, AI4Imaging).

Frequently Asked Questions

What is radiomics in medical imaging?

Radiomics involves high-throughput extraction of quantitative features from medical images to characterize tumor intensity, texture, and shape. "Radiomics: Images Are More than Pictures, They Are Data" (Gillies et al., 2015) notes that advances in pattern recognition have facilitated this process from exponentially growing datasets. These features support machine learning models for predictive tasks in cancer imaging.

How does machine learning integrate with radiomics?

Machine learning algorithms process high-dimensional radiomics features for classification and predictive modeling. "A survey on deep learning in medical image analysis" (Litjens et al., 2017) reviews deep learning applications in medical image analysis, including segmentation relevant to radiomics pipelines. Recent preprints like "Machine Learning–Driven Radiomics for Glioma Diagnosis" combine these for PET-based glioma classification.

What tools support radiomics feature extraction?

Pyradiomics is an open-source Python package for extracting radiomics features from 2D and 3D images and binary masks. "3D Slicer as an image computing platform for the Quantitative Imaging Network" (Fedorov et al., 2012) provides an image computing platform supporting such quantitative analysis. These tools establish reference standards for reproducible radiomic analysis.

What are current applications in cancer imaging?

Applications include tumor staging and response evaluation, as in "New response evaluation criteria in solid tumours: Revised RECIST guideline (version 1.1)" (Eisenhauer et al., 2008) with 28,308 citations. Machine learning enhances segmentation in "UNet++: A Nested U-Net Architecture for Medical Image Segmentation" (Zhou et al., 2018). Preprints address glioma diagnosis and radiation therapy via radiomics.

What guidelines exist for robust radiomics studies?

The Radiomics Quality Score (RQS) 2.0 assesses study rigor for clinical translation in precision oncology. "Robust radiomics: a review of guidelines for radiomics in medical imaging" summarizes pipelines combining quantitative analysis with AI. These promote reproducibility in medical imaging research.

What is the state of reproducibility in radiomics?

Challenges include reproducibility and accessibility, addressed by "Open-radiomics: a collection of standardized datasets and a technical protocol for reproducible radiomics machine learning pipelines." This provides datasets and protocols for feature extraction, selection, and classification. Forks like bhklab/pyradiomics-fork maintain open-source standards.

Open Research Questions

  • ? How can radiomics features be standardized across multi-center datasets to improve reproducibility in machine learning models?
  • ? What pipelines best integrate deep learning segmentation with radiomics for real-time clinical predictive modeling?
  • ? Which radiomics features most accurately predict treatment response beyond RECIST criteria in solid tumors?
  • ? How do guidelines like RQS 2.0 influence the clinical translation of radiomics in precision oncology?
  • ? What effects do imaging protocols have on the robustness of machine learning-driven radiomics for glioma diagnosis?

Research Radiomics and Machine Learning in Medical Imaging with AI

PapersFlow provides specialized AI tools for Medicine researchers. Here are the most relevant for this topic:

See how researchers in Health & Medicine use PapersFlow

Field-specific workflows, example queries, and use cases.

Health & Medicine Guide

Start Researching Radiomics and Machine Learning in Medical Imaging with AI

Search 474M+ papers, run AI-powered literature reviews, and write with integrated citations — all in one workspace.

See how PapersFlow works for Medicine researchers