Copycat

Dietz, Brett W.; Winston, Lisa G; Koehler, Jane E.; Margaretten, Mary

doi:10.1056/nejmcps2108885

Cited by 7 publications

(7 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…-Disease with characteristic symptoms -Acute intermittent porphyria [21] -Di cult-to-diagnose disease with few characteristic symptoms -Acute HIV-1 infection [22] -Case with diagnostic errors -Subacute bacterial endocarditis caused by bartonella [23] -DDx performance using the Gumbel Approximate NDCG loss -Acute intermittent porphyria [21] -Acute HIV-1 infection [22] In the interest of fairness, these cases were not used when training the model.…”

Section: Evaluation Methodsmentioning

confidence: 99%

“…The nal con rmed disease of the case was subacute bacterial endocarditis caused by bartonella. [23] The title of the case report is "Copycat." In this case, this patient had a history of HCV infection.…”

Section: Case With Diagnostic Errorsmentioning

confidence: 99%

“…Target diseases HIV-1 AVM Case: subacute bacterial endocarditis caused by bartonella [23] Loss functions: Approximate NDCG loss (A-NDCG)…”

Section: Inputted Symptoms Rankingmentioning

confidence: 99%

See 2 more Smart Citations

Design, Implementation, and Evaluation of the Computer-aided Clinical Decision Support System based on Learning-to-Rank: Collaboration between physicians and machine learning in the differential diagnosis process

Miyachi¹,

Ishii²,

Torigoe

2022

Preprint

View full text Add to dashboard Cite

Background We are researching, developing, and publishing the Clinical Decision Support System based on Learning-to-Rank. The main objectives are 1) Supporting differential diagnoses by internists and general practitioners and 2) Preventing diagnostic errors by physicians. The main features are that "A physician inputs a patient's symptoms, findings, and test results to the system, and the system outputs a ranking list of possible diseases." Method The software libraries for machine learning and artificial intelligence are TensorFlow and TensorFlow Ranking. The prediction algorithm is Learning-to-Rank., with a listwise approach. The ranking metric is Normalized Discounted Cumulative Gain (NDCG). The loss functions are Approximate NDCG (A-NDCG) and Gumbel Approximate NDCG (G-A-NDCG). We evaluated the machine learning performance on k-fold cross-validation. We evaluated the differential diagnosis performance with actual cases. Results The machine learning performance of our system was much higher than that of the conventional system. The machine learning performance using G-A-NDCG was slightly higher than that of A-NDCG. The differential diagnosis performance of our system was much higher than that of the conventional system. We have shown that the Clinical Decision Support System prevents physicians' diagnostic errors due to confirmation bias. Conclusions We have demonstrated that the Clinical Decision Support System is useful for supporting differential diagnoses and preventing diagnostic errors. We believe that Differential diagnosis by physicians and Learning-to-Rank by machine has a high affinity. We found that Information Retrieval and Clinical Decision Support Systems have much in common (Target data, Learning-to-Rank, etc.). We believe that Clinical Decision Support Systems have the potential to support 1) recall of rare diseases, 2) differential diagnoses for difficult-to-diagnose diseases, and 3) prevention of diagnostic errors. We also believe that our system has the potential to evolve into an Explainable Clinical Decision Support System.

show abstract

Section: Evaluation Methodsmentioning

confidence: 99%

Section: Case With Diagnostic Errorsmentioning

confidence: 99%

See 1 more Smart Citation

Design, Implementation, and Evaluation of the Computer-aided Clinical Decision Support System based on Learning-to-Rank: Collaboration between physicians and machine learning in the differential diagnosis process

Miyachi¹,

Ishii²,

Torigoe

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…To further verify the malignant traits of the two metabolic subtypes obtained by bulk RNA-seq analysis, copycat (Dietz et al, 2021) was used to calculate the large-scale chromosomal copy Frontiers in Genetics frontiersin.org number variation (CNV) in each fine cell type based on the scRNAseq data of six kinds of immune cells, so as to distinguish malignant cells from non-malignant cells in each sample. There was an obvious difference in the proportion of malignant cells and non-malignant cells between cluster 1 and cluster 2.…”

Section: Composition and Metabolism Of Malignant Cellsmentioning

confidence: 99%

Combining bulk RNA-sequencing and single-cell RNA-sequencing data to reveal the immune microenvironment and metabolic pattern of osteosarcoma

Huang

Wang²,

Yin³

et al. 2022

Front. Genet.

View full text Add to dashboard Cite

Background: Osteosarcoma (OS) is a kind of solid tumor with high heterogeneity at tumor microenvironment (TME), genome and transcriptome level. In view of the regulatory effect of metabolism on TME, this study was based on four metabolic models to explore the intertumoral heterogeneity of OS at the RNA sequencing (RNA-seq) level and the intratumoral heterogeneity of OS at the bulk RNA-seq and single cell RNA-seq (scRNA-seq) level.Methods: The GSVA package was used for single-sample gene set enrichment analysis (ssGSEA) analysis to obtain a glycolysis, pentose phosphate pathway (PPP), fatty acid oxidation (FAO) and glutaminolysis gene sets score. ConsensusClusterPlus was employed to cluster OS samples downloaded from the Target database. The scRNA-seq and bulk RNA-seq data of immune cells from GSE162454 dataset were analyzed to identify the subsets and types of immune cells in OS. Malignant cells and non-malignant cells were distinguished by large-scale chromosomal copy number variation. The correlations of metabolic molecular subtypes and immune cell types with four metabolic patterns, hypoxia and angiogenesis were determined by Pearson correlation analysis.Results: Two metabolism-related molecular subtypes of OS, cluster 1 and cluster 2, were identified. Cluster 2 was associated with poor prognosis of OS, active glycolysis, FAO, glutaminolysis, and bad TME. The identified 28608 immune cells were divided into 15 separate clusters covering 6 types of immune cells. The enrichment scores of 5 kinds of immune cells in cluster-1 and cluster-2 were significantly different. And five kinds of immune cells were significantly correlated with four metabolic modes, hypoxia and angiogenesis. Of the 28,608 immune cells, 7617 were malignant cells. The four metabolic patterns of malignant cells were significantly positively correlated with hypoxia and negatively correlated with angiogenesis.Conclusion: We used RNA-seq to reveal two molecular subtypes of OS with prognosis, metabolic pattern and TME, and determined the composition and metabolic heterogeneity of immune cells in OS tumor by bulk RNA-seq and single-cell RNA-seq.

show abstract

“…The nal con rmed disease of the case was subacute bacterial endocarditis caused by bartonella. [Dietz et al, 2021] The title of the case report is "Copycat." In this case, this patient had a history of HCV infection.…”

Section: Subacute Bacterial Endocarditis Caused By Bartonellamentioning

confidence: 99%

Design, Implementation, and Evaluation of the Computer-aided Clinical Decision Support System based on Learning-to-Rank: Collaboration between physicians and machine learning in the differential diagnosis process

Miyachi¹,

Ishii²,

Torigoe³

2022

Preprint

View full text Add to dashboard Cite

OBJECTIVES We are developing the Clinical Decision Support System (CDSS) based on Learning-to-Rank (LTR). The main objectives are 1) Supporting differential diagnoses by internists and general practitioners and 2) Preventing diagnostic errors by physicians. The main features are that "A physician inputs a patient's symptoms, findings, and test results to the system, and the system outputs a ranking list of possible diseases." METHOD The software libraries for machine learning and artificial intelligence are TensorFlow and TensorFlow Ranking. The prediction algorithm is LTR with a listwise approach. The ranking metric is NDCG. The loss functions are Approximate NDCG (A-NDCG) and Gumbel Approximate NDCG (G-A-NDCG). We evaluated Machine Learning (ML) performance and Differential Diagnosis (DDx) performance with actual cases. RESULTS ML performance of our system was much higher than that of the conventional system. ML performance using G-A-NDCG was slightly higher than that of A-NDCG. DDx performance of our system was much higher than that of the conventional system. We have shown that CDSS prevents physicians' diagnostic errors due to confirmation bias. CONCLUSIONS We have demonstrated that the CDSS is useful for supporting differential diagnoses and preventing diagnostic errors. We believe that DDx by physicians and LTR have a high affinity. We found that Information Retrieval (IR) and Clinical Decision Support System (CDSS) have much in common (target data, LTR, etc.). We believe that CDSS has the potential to support 1) recall of rare diseases, 2) differential diagnoses for difficult-to-diagnose diseases, and 3) prevention of diagnostic errors. We also believe that our system has the potential for evolution to an Explainable Clinical Decision Support System (X-CDSS).

show abstract

Copycat

Cited by 7 publications

References 10 publications

Design, Implementation, and Evaluation of the Computer-aided Clinical Decision Support System based on Learning-to-Rank: Collaboration between physicians and machine learning in the differential diagnosis process

Design, Implementation, and Evaluation of the Computer-aided Clinical Decision Support System based on Learning-to-Rank: Collaboration between physicians and machine learning in the differential diagnosis process

Combining bulk RNA-sequencing and single-cell RNA-sequencing data to reveal the immune microenvironment and metabolic pattern of osteosarcoma

Design, Implementation, and Evaluation of the Computer-aided Clinical Decision Support System based on Learning-to-Rank: Collaboration between physicians and machine learning in the differential diagnosis process

Contact Info

Product

Resources

About