Towards a safe and efficient clinical implementation of machine learning in radiation oncology by exploring model interpretability, explainability and data-model dependency
2022/04/14 by Ana María Barragán Montero, Adrien Bibal, Margerie Huet Dastarac +10 · 1 voice
Computer Science · Medicine · #Artificial Intelligence in Healthcare and Education #Explainable Artificial Intelligence (XAI) #Radiomics and Machine Learning in Medical Imaging
paper · pdf · doi:10.1088/1361-6560/ac678a
openalex publication_date 2022/04/14 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/01
Abstract
The interest in machine learning (ML) has grown tremendously in recent years, partly due to the performance leap that occurred with new techniques of deep learning, convolutional neural networks for images, increased computational power, and wider availability of large datasets. Most fields of medicine follow that popular trend and, notably, radiation oncology is one of those that are at the forefront, with already a long tradition in using digital images and fully computerized workflows. ML models are driven by data, and in contrast with many statistical or physical models, they can be very large and complex, with countless generic parameters. This inevitably raises two questions, namely, the tight dependence between the models and the datasets that feed them, and the interpretability of the models, which scales with its complexity. Any problems in the data used to train the model will be later reflected in their performance. This, together with the low interpretability of ML models, makes their implementation into the clinical workflow particularly difficult. Building tools for risk assessment and quality assurance of ML models must involve then two main points: interpretability and data-model dependency. After a joint introduction of both radiation oncology and ML, this paper reviews the main risks and current solutions when applying the latter to workflows in the former. Risks associated with data and models, as well as their interaction, are detailed. Next, the core concepts of interpretability, explainability, and data-model dependency are formally defined and illustrated with examples. Afterwards, a broad discussion goes through key applications of ML in workflows of radiation oncology as well as vendors' perspectives for the clinical implementation of ML.
Citations
- Intriguing properties of neural networks
- Unsupervised Domain Adaptation by Backpropagation
- Intelligible Models for HealthCare
- Textural features corresponding to textural properties
- Textural Features for Image Classification
- Zero-Shot Learning Through Cross-Modal Transfer
- Neural Machine Translation by Jointly Learning to Align and Translate
- Representation Learning: A Review and New Perspectives
- A Survey on Transfer Learning
- Towards A Rigorous Science of Interpretable Machine Learning
- One-Shot Imitation Learning
- Prototypical Networks for Few-shot Learning
- Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
- Delineation of the primary tumour Clinical Target Volumes (CTV-P) in laryngeal, hypopharyngeal, oropharyngeal and oral cavity squamous cell carcinoma: AIRO, CACA, DAHANCA, EORTC, GEORCC, GORTEC, HKNPCSG, HNCIG, IAG-KHT, LPRHHT, NCIC CTG, NCRI, NRG Oncology, PHNS, SBRT, SOMERA, SRO, SSHNO, TROG consensus guidelines
- Dose evaluation of fast synthetic-CT generation using a generative adversarial network for general pelvis MR-only radiotherapy
- Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
- Adversarial attacks on medical machine learning
- BatchBALD: Efficient and Diverse Batch Acquisition for Deep Bayesian Active Learning
- Weight Uncertainty in Neural Networks
- A Survey of Methods for Explaining Black Box Models
- Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps
- On a Method to Measure Supervised Multiclass Model’s Interpretability: Application to Degradation Diagnosis (Short Paper)
- On Identifiability in Transformers
- ALBERT: A Lite BERT for Self-supervised Learning of Language Representations
- nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation
- Shortcut learning in deep neural networks
- Self-Supervised Visual Feature Learning With Deep Neural Networks: A Survey
- 3D Self-Supervised Methods for Medical Imaging
- Generalizing from a Few Examples
- Opportunities and Challenges in Explainable Artificial Intelligence (XAI): A Survey
- Underspecification Presents Challenges for Credibility in Modern Machine Learning
- A review of uncertainty quantification in deep learning: Techniques, applications and challenges
- A Survey on Multi-Task Learning
- A review of medical image data augmentation techniques for deep learning applications
- Text Data Augmentation for Deep Learning
- Exploiting Generative AI to Scale up Intelligent Tutoring Systems
Discussions