Background: Large language models such as ChatGPT can produce increasingly realistic text, with unknown information on the accuracy and integrity of using these models in scientific writing. Methods: We gathered ten research abstracts from five high impact factor medical journals (n=50) and asked ChatGPT to generate research abstracts based on their titles and journals. We evaluated the abstracts using an artificial intelligence (AI) output detector, plagiarism detector, and had blinded human reviewers try to distinguish whether abstracts were original or generated. Results: All ChatGPT-generated abstracts were written clearly but only 8% correctly followed the specific journal's formatting requirements. Most generated abstracts were detected using the AI output detector, with scores (higher meaning more likely to be generated) of median [interquartile range] of 99.98% [12.73, 99.98] compared with very low probability of AI-generated output in the original abstracts of 0.02% [0.02, 0.09]. The AUROC of the AI output detector was 0.94. Generated abstracts scored very high on originality using the plagiarism detector (100% [100, 100] originality). Generated abstracts had a similar patient cohort size as original abstracts, though the exact numbers were fabricated. When given a mixture of original and general abstracts, blinded human reviewers correctly identified 68% of generated abstracts as being generated by ChatGPT, but incorrectly identified 14% of original abstracts as being generated. Reviewers indicated that it was surprisingly difficult to differentiate between the two, but that the generated abstracts were vaguer and had a formulaic feel to the writing. Conclusion: ChatGPT writes believable scientific abstracts, though with completely generated data. These are original without any plagiarism detected but are often identifiable using an AI output detector and skeptical human reviewers. Abstract evaluation for journals and medical conferences must adapt policy and practice to maintain rigorous scientific standards; we suggest inclusion of AI output detectors in the editorial process and clear disclosure if these technologies are used. The boundaries of ethical and acceptable use of large language models to help scientific writing remain to be determined.

Large language models such as ChatGPT can produce increasingly realistic text, with unknown information on the accuracy and integrity of using these models in scientific writing. We gathered fifth research abstracts from five high-impact factor medical journals and asked ChatGPT to generate research abstracts based on their titles and journals. Most generated abstracts were detected using an AI output detector, ‘GPT-2 Output Detector’, with % ‘fake’ scores (higher meaning more likely to be generated) of median [interquartile range] of 99.98% ‘fake’ [12.73%, 99.98%] compared with median 0.02% [IQR 0.02%, 0.09%] for the original abstracts. The AUROC of the AI output detector was 0.94. Generated abstracts scored lower than original abstracts when run through a plagiarism detector website and iThenticate (higher scores meaning more matching text found). When given a mixture of original and general abstracts, blinded human reviewers correctly identified 68% of generated abstracts as being generated by ChatGPT, but incorrectly identified 14% of original abstracts as being generated. Reviewers indicated that it was surprisingly difficult to differentiate between the two, though abstracts they suspected were generated were vaguer and more formulaic. ChatGPT writes believable scientific abstracts, though with completely generated data. Depending on publisher-specific guidelines, AI output detectors may serve as an editorial tool to help maintain scientific standards. The boundaries of ethical and acceptable use of large language models to help scientific writing are still being discussed, and different journals and conferences are adopting varying policies.

A model’s ability to express its own predictive uncertainty is an essential attribute for maintaining clinical user confidence as computational biomarkers are deployed into real-world medical settings. In the domain of cancer digital histopathology, we describe a clinically-oriented approach to uncertainty quantification for whole-slide images, estimating uncertainty using dropout and calculating thresholds on training data to establish cutoffs for low- and high-confidence predictions. We train models to identify lung adenocarcinoma vs. squamous cell carcinoma and show that high-confidence predictions outperform predictions without uncertainty, in both cross-validation and testing on two large external datasets spanning multiple institutions. Our testing strategy closely approximates real-world application, with predictions generated on unsupervised, unannotated slides using predetermined thresholds. Furthermore, we show that uncertainty thresholding remains reliable in the setting of domain shift, with accurate high-confidence predictions of adenocarcinoma vs. squamous cell carcinoma for out-of-distribution, non-lung cancer cohorts.

The histopathological phenotype of tumors reflects the underlying genetic makeup. Deep learning can predict genetic alterations from pathology slides, but it is unclear how well these predictions generalize to external datasets. We performed a systematic study on Deep-Learning-based prediction of genetic alterations from histology, using two large datasets of multiple tumor types. We show that an analysis pipeline that integrates self-supervised feature extraction and attention-based multiple instance learning achieves a robust predictability and generalizability.

Artificial intelligence methods including deep neural networks (DNN) can provide rapid molecular classification of tumors from routine histology with accuracy that matches or exceeds human pathologists. Discerning how neural networks make their predictions remains a significant challenge, but explainability tools help provide insights into what models have learned when corresponding histologic features are poorly defined. Here, we present a method for improving explainability of DNN models using synthetic histology generated by a conditional generative adversarial network (cGAN). We show that cGANs generate high-quality synthetic histology images that can be leveraged for explaining DNN models trained to classify molecularly-subtyped tumors, exposing histologic features associated with molecular state. Fine-tuning synthetic histology through class and layer blending illustrates nuanced morphologic differences between tumor subtypes. Finally, we demonstrate the use of synthetic histology for augmenting pathologist-in-training education, showing that these intuitive visualizations can reinforce and improve understanding of histologic manifestations of tumor biology.