AI-based pathology predicts origins for cancers of unknown primary
By Ming Y. Lu, Tiffany Y. Chen, Drew F. K. Williamson, Melissa Zhao, Maha Shady, Jana Lipková, Faisal Mahmood. Nature, 2020
622 citations recorded by the source corpus.
Cancer of unknown primary (CUP) origin is an enigmatic group of diagnoses in which the primary anatomical site of tumour origin cannot be determined1,2. This poses a considerable challenge, as modern therapeutics are predominantly specific to the primary tumour3. Recent research has focused on using genomics and transcriptomics to identify the origin of a tumour4–9. However, genomic testing is not always performed and lacks clinical penetration in low-resource settings. Here, to overcome these challenges, we present a deep-learning-based algorithm—Tumour Origin Assessment via Deep Learning (TOAD)—that can provide a differential diagnosis for the origin of the primary tumour using routinely acquired histology slides. We used whole-slide images of tumours with known primary origins to train a model that simultaneously identifies the tumour as primary or metastatic and predicts its site of origin. On our held-out test set of tumours with known primary origins, the model achieved a top-1 accuracy of 0.83 and a top-3 accuracy of 0.96, whereas on our external test set it achieved top-1 and top-3 accuracies of 0.80 and 0.93, respectively. We further curated a dataset of 317 cases of CUP for which a differential diagnosis was assigned. Our model predictions resulted in concordance for 61% of cases and a top-3 agreement of 82%. TOAD can be used as an assistive tool to assign a differential diagnosis to complicated cases of metastatic tumours and CUPs and could be used in conjunction with or in lieu of ancillary tests and extensive diagnostic work-ups to reduce the occurrence of CUP. A deep-learning-based algorithm uses routinely acquired histology slides to provide a differential diagnosis for the origin of the primary tumour for complicated cases of metastatic tumours and cancers of unknown primary origin.
Subfields: Oncology, Otorhinolaryngology, Periodontics
View Paper: https://www.semanticscholar.org/paper/6edc9d0937b450d6facdbf7bbf1131bbb708fa2a
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Oncology
Citations in the corpus
- Multimodal deep learning for biomedical data fusion: a review (2022, background): “[66] showed that concatenating biological sex of the patient as a covariate with features learned from WSIs improved the performance in predicting primary sites of cancers of unknown primary.”
- From patterns to patients: Advances in clinical machine learning for cancer diagnosis, prognosis, and treatment. (2023, methodology): “trained a ResNet-50-based model on H&E images to identify a tumor as primary or metastatic and predict its site of origin across 18 different primary origins, with top-3 prediction accuracy on an external set exceeding 90%.(15) The model incorporated attention after the CNN layers, which identified ”
- Deep learning in cancer diagnosis, prognosis and treatment selection (2021, methodology): “(WSIs) was able to simultaneously predict metastasis status and origin of 18 tumour types [58].”
- Scaling Vision Transformers to Gigapixel Images via Hierarchical Self-Supervised Learning (2022, neutral): “- subtype, grade, stage, survival, origin) [12, 19, 37, 38, 52, 53, 68, 70, 85].”
- Artificial intelligence in histopathology: enhancing cancer research and clinical oncology (2022, background): “Additionally, digital pathology is capable of classifying cancers of unknown primary origin, which can be difficult to diagnose and treat 31 .”
- Algorithm fairness in artificial intelligence for medicine and healthcare (2023, background): “Although the sex of a patient can be viewed as a sensitive attribute, not including this information may result in unusual diagnoses, such as predicting the prostate as the origin of a tumour in cases of cancer of unknown primary in women 165 .”
- Multimodal Co-Attention Transformer for Survival Prediction in Gigapixel Whole Slide Images (2021, background): “Recent work has demonstrated remarkable progress in using multiple Instance Learning (MIL) and other set-based deep learning approaches for learning tasks in gigapixel images [23, 5, 52, 51, 62, 69, 37, 36].”
- Artificial intelligence for digital and computational pathology (2023, methodology): “Alternatively, WSI classification can be defined with multiple instance learning (MIL) 31,32 , in which a single supervisory label is provided to the set of patches constituting the WSI and only a subset of the patches is assumed to correspond to that label.”