Evaluation and accurate diagnoses of pediatric diseases using artificial intelligence

Liang, Huiying; Tsui, Brian; Ni, Hao; Valentim, Carolina C S; Baxter, Sally L.; Liu, Guangjian; Cai, Wenjia; Kermany, Daniel; Sun, Xin; Chen, Jiancong; He, Liang; Zhu, Jie; Pin, Tian; Shao, Hua; Zheng, Lianghong; Hou, Rui; Hewett, Sierra; Li, Gen; Liang, Ping; Zang, Xuan; Zhang, Zhiqi; Pan, Liyan; Cai, H.; Ling, Rujuan; Li, Shuhua; Cui, Yongwang; Tang, Shusheng; Yu, Hong; Huang, Xiaoyan; He, Wan-er; Liang, Wenqing; Zhang, Qing; Jiang, Jianmin; Yu, Wei; Gao, Jianqun; Ou, Wanxing; Deng, Yingmin; Hou, Qiaozhen; Wang, Bei; Yao, Cuichan; Liang, Yan; Zhang, Shu; Duan, Yaou; Zhang, Runze; Gibson, Sarah; Zhang, Charlotte L; Li, Oulan; Zhang, Edward D.; Karin, Gabriel; Nguyen, Nathan; Wu, Xiaokang; Wen, Cindy; Xu, Jie; Xu, W.; Wang, Bochu; Wang, Winston; Li, Jing; Pizzato, Bianca Ribeiro; Bao, Caroline; Xiang, Daoman; He, Wanting; He, Suiqin; Zhou, Yugui; Haw, Weldon W; Goldbaum, Michael H.; Tremoulet, Adriana H.; Hsu, Chun‐Nan; Carter, Hannah; Zhu, Long; Zhang, Kang; Xia, Huimin

doi:10.1038/s41591-018-0335-9

Cited by 486 publications

(320 citation statements)

References 19 publications

Supporting

Mentioning

273

Contrasting

Unclassified

Order By: Relevance

“…(2) Measure precision/recall rates directly comparable with concordance approaches Using only the labeled calls for calculating precision and recall rates is a standard practice for measuring the performance of machine learning classifiers (Saito and Rehmsmeier 2015;Liang et al 2019). Since the original CNV calling algorithms do not use such labels to make CNV predictions, the recall rates for the naïve concordance-based methods could only be calculated as the number of calls predicted from individual callers or concordant calls from two or more callers divided by the total number of validated CNVs (Supplemental Fig.…”

Section: (1) Performance Evaluation Of Cn-learnmentioning

confidence: 99%

A machine-learning approach for accurate detection of copy number variants from exome sequencing

et al. 2019

View full text Add to dashboard Cite

Copy number variants (CNVs) are a major cause of several genetic disorders, making their detection an essential component of genetic analysis pipelines. Current methods for detecting CNVs from exome-sequencing data are limited by high false-positive rates and low concordance because of inherent biases of individual algorithms. To overcome these issues, calls generated by two or more algorithms are often intersected using Venn diagram approaches to identify “high-confidence” CNVs. However, this approach is inadequate, because it misses potentially true calls that do not have consensus from multiple callers. Here, we present CN-Learn, a machine-learning framework that integrates calls from multiple CNV detection algorithms and learns to accurately identify true CNVs using caller-specific and genomic features from a small subset of validated CNVs. Using CNVs predicted by four exome-based CNV callers (CANOES, CODEX, XHMM, and CLAMMS) from 503 samples, we demonstrate that CN-Learn identifies true CNVs at higher precision (∼90%) and recall (∼85%) rates while maintaining robust performance even when trained with minimal data (∼30 samples). CN-Learn recovers twice as many CNVs compared to individual callers or Venn diagram–based approaches, with features such as exome capture probe count, caller concordance, and GC content providing the most discriminatory power. In fact, ∼58% of all true CNVs recovered by CN-Learn were either singletons or calls that lacked support from at least one caller. Our study underscores the limitations of current approaches for CNV identification and provides an effective method that yields high-quality CNVs for application in clinical diagnostics.

show abstract

Section: (1) Performance Evaluation Of Cn-learnmentioning

confidence: 99%

A machine-learning approach for accurate detection of copy number variants from exome sequencing

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Applications of AI and deep learning argued to be useful tools in assisting diagnosis and treatment decision making [10][11]. There were studies which promoted disease detection through AI models [12][13][14][15].…”

Section: The Centers For Disease Control and Prevention (Cdc) And Wormentioning

confidence: 99%

Identification of COVID-19 can be quicker through artificial intelligence framework using a mobile phone–based survey when cities and towns are under quarantine

Rao

Vázquez

2020

Infect. Control Hosp. Epidemiol.

336

139

View full text Add to dashboard Cite

show abstract

“…For example, in radiology, natural‐language processing was used to automatically notate whether a certain condition or finding is mentioned within the text of the report . In another example, Liang and others used natural‐language processing to allow the use of unstructured information from EHRs for the development of a deep‐learning model for automatic pediatric diagnoses that surpassed the accuracy of junior, but not senior, physicians . Finally, the data extracted using natural‐language processing might be required to identify patients eligible for inclusion in cohorts for observational research …”

Section: Advances In Pediatric Data Collectionmentioning

confidence: 99%

“…16 In another example, Liang and others used natural-language processing to allow the use of unstructured information from EHRs for the development of a deep-learning model for automatic pediatric diagnoses that surpassed the accuracy of junior, but not senior, physicians. 17 Finally, the data extracted using natural-language processing might be required to identify patients eligible for inclusion in cohorts for observational research. 15 Although effectiveness research with real-world data can be problematic due to the difficulty in controlling for confounding variables and nonrandomized treatment decisions, real-world data offer many other opportunities.…”

Section: • Limited Control • Complex Data Analysis • No Expert Supervmentioning

confidence: 99%

Beyond the Randomized Clinical Trial: Innovative Data Science to Close the Pediatric Evidence Gap

Goulooze

Zwep

Vogt

et al. 2020

Clin Pharma and Therapeutics

View full text Add to dashboard Cite

Despite the application of advanced statistical and pharmacometric approaches to pediatric trial data, a large pediatric evidence gap still remains. Here, we discuss how to collect more data from children by using real‐world data from electronic health records, mobile applications, wearables, and social media. The large datasets collected with these approaches enable and may demand the use of artificial intelligence and machine learning to allow the data to be analyzed for decision making. Applications of this approach are presented, which include the prediction of future clinical complications, medical image analysis, identification of new pediatric end points and biomarkers, the prediction of treatment nonresponders, and the prediction of placebo‐responders for trial enrichment. Finally, we discuss how to bring machine learning from science to pediatric clinical practice. We conclude that advantage should be taken of the current opportunities offered by innovations in data science and machine learning to close the pediatric evidence gap.

show abstract

Evaluation and accurate diagnoses of pediatric diseases using artificial intelligence

Cited by 486 publications

References 19 publications

A machine-learning approach for accurate detection of copy number variants from exome sequencing

A machine-learning approach for accurate detection of copy number variants from exome sequencing

Identification of COVID-19 can be quicker through artificial intelligence framework using a mobile phone–based survey when cities and towns are under quarantine

Beyond the Randomized Clinical Trial: Innovative Data Science to Close the Pediatric Evidence Gap

Contact Info

Product

Resources

About