Development and Validation of a Model to Predict Anterior Segment Vision-Threatening Eye Disease Using Primary Care Clinical Notes

Singh, Karandeep; Thibodeau, Alexa; Niziol, Leslie M.; Nakai, Tejpreet K.; Bixler, Jill; Khan, Mariam; Woodward, Maria A.

doi:10.1097/ico.0000000000002877

Cited by 2 publications

(7 citation statements)

References 21 publications

(26 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Of the 15 EHR papers, 13 included the development of a prediction model [19][20][21][22][23][24][25][26][27][28][29][30][31]. In Table 1, the data extraction per paper can be found.…”

Section: Overview Of the Ehr-based Papersmentioning

confidence: 99%

“…Predictors included findings from clinical examination (n = 6) [20,26,27,29,32,33], laboratory results (n = 5) [22,23,26,29,33] and medication (n = 4) [20,22,25,30]. All models used structured data.…”

Section: Overview Of the Ehr-based Papersmentioning

confidence: 99%

“…One paper had a prospective approach and tested the diagnostic performance of a prediction model for Asthma and COPD [32]. Ten papers were published after 2020 [19,[22][23][24][25]27,28,30,32,33]. Most data sources used in the studies originated from Europe (n = 8) [19][20][21][22][23]25,28,32], followed by North America (n = 5) [24,26,29,30,32,33] and Asia (n = 2) [27,31].…”

Section: Overview Of the Ehr-based Papersmentioning

confidence: 99%

“…Ten papers were published after 2020 [19,[22][23][24][25]27,28,30,32,33]. Most data sources used in the studies originated from Europe (n = 8) [19][20][21][22][23]25,28,32], followed by North America (n = 5) [24,26,29,30,32,33] and Asia (n = 2) [27,31].…”

Section: Overview Of the Ehr-based Papersmentioning

confidence: 99%

“…The main reasons for the high risk of bias in this domain were the insufficient number of participants with the outcome (n = 5) [19,25,27,30,32] and irrelevant model performance measures that were used to evaluate the model (n = 2) [29,32]. The main reasons for an unclear risk of bias in the analysis domain were lack of clarity on how missing data were handled (n = 10) [19][20][21]23,24,[28][29][30][31]33], and on how the predictors and their assigned weights in the final model correspond to results from the reported multivariate analysis (n = 9) [19,21,22,[26][27][28][29][30][31]. Even though measures of calibration are not part of the signalling questions of the PROBAST, we noticed that only 4 papers [23,24,30,33] used calibration to assess the performance of the model.…”

Section: Risk Of Biasmentioning

confidence: 99%

See 4 more Smart Citations

Diagnostic prediction models for primary care, based on artificial intelligence and electronic health records: a systematic review (Preprint)

Hunik,

Chaabouni,

van Laarhoven

et al. 2024

Preprint

View full text Add to dashboard Cite

BACKGROUND Artificial intelligence (AI) based diagnostic prediction models could aid primary care (PC) in decision making for faster and more accurate diagnoses. AI has the potential to transform electronic health records (EHR) data into valuable diagnostic prediction models. Different prediction models based on EHR have been developed. However, there are currently no systematic reviews that evaluate AI-based diagnostic prediction models for PC using EHR data. OBJECTIVE To provide an overview of diagnostic prediction models based on AI and EHR in primary care and to evaluate the content of each model, including risk of bias and applicability. METHODS This systematic review was performed according to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. MEDLINE, EMBASE, Web of Science and Cochrane were searched. We included observational and intervention studies using AI and primary care EHRs and developing or testing a diagnostic prediction model for health conditions. Two independent reviewers used a standardised data extraction form. Risk of bias and applicability were assessed using PROBAST (Prediction model Risk Of Bias ASsessment Tool). RESULTS From 10,657 retrieved records, a total of 15 papers were selected. Most EHR papers focused on one chronic healthcare condition (n=11). From the 15 papers, 13 described a study that developed a diagnostic prediction model and 2 described a study that externally validated and tested the model in a primary care setting. Studies used a variety of AI techniques. The predictors used to develop the model were all registered in the EHR. We found no papers with a low risk of bias, high risk of bias was found in 9 papers. Biases covered an unjustified small sample size (n=5), not excluding predictors from the outcome definition (n=2) and the inappropriate evaluation of the performance measures (n=2). Unclear risk of bias was found in 6 papers, as no information was provided on the handling of missing data (n=10) and no results were reported from the multivariate analysis (n=9). Applicability was unclear in 10 papers, mainly due to lack of clarity in reporting the time interval between outcomes and predictors. CONCLUSIONS Most AI-based diagnostic prediction models based on EHR data in primary care focused on one chronic condition. Only two papers tested the model in a primary care setting. The lack of sufficiently described methods led to a high risk of bias. Our findings highlight that the currently available diagnostic prediction models are not yet ready for clinical implementation in primary care.

show abstract

“…Of the 15 EHR papers, 13 included the development of a prediction model [19][20][21][22][23][24][25][26][27][28][29][30][31]. In Table 1, the data extraction per paper can be found.…”

Section: Overview Of the Ehr-based Papersmentioning

confidence: 99%

Section: Overview Of the Ehr-based Papersmentioning

confidence: 99%

Section: Overview Of the Ehr-based Papersmentioning

confidence: 99%

Section: Overview Of the Ehr-based Papersmentioning

confidence: 99%

Section: Risk Of Biasmentioning

confidence: 99%

See 3 more Smart Citations

Diagnostic prediction models for primary care, based on artificial intelligence and electronic health records: a systematic review (Preprint)

Hunik,

Chaabouni,

van Laarhoven

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

Exploring large language model for next generation of artificial intelligence in ophthalmology

Jin,

Yuan,

et al. 2023

Front. Med.

View full text Add to dashboard Cite

In recent years, ophthalmology has advanced significantly, thanks to rapid progress in artificial intelligence (AI) technologies. Large language models (LLMs) like ChatGPT have emerged as powerful tools for natural language processing. This paper finally includes 108 studies, and explores LLMs’ potential in the next generation of AI in ophthalmology. The results encompass a diverse range of studies in the field of ophthalmology, highlighting the versatile applications of LLMs. Subfields encompass general ophthalmology, retinal diseases, anterior segment diseases, glaucoma, and ophthalmic plastics. Results show LLMs’ competence in generating informative and contextually relevant responses, potentially reducing diagnostic errors and improving patient outcomes. Overall, this study highlights LLMs’ promising role in shaping AI’s future in ophthalmology. By leveraging AI, ophthalmologists can access a wealth of information, enhance diagnostic accuracy, and provide better patient care. Despite challenges, continued AI advancements and ongoing research will pave the way for the next generation of AI-assisted ophthalmic practices.

show abstract

Development and Validation of a Model to Predict Anterior Segment Vision-Threatening Eye Disease Using Primary Care Clinical Notes

Abstract: Supplemental Digital Content is Available in the Text.

Cited by 2 publications

References 21 publications

Diagnostic prediction models for primary care, based on artificial intelligence and electronic health records: a systematic review (Preprint)

Diagnostic prediction models for primary care, based on artificial intelligence and electronic health records: a systematic review (Preprint)

Exploring large language model for next generation of artificial intelligence in ophthalmology

Contact Info

Product

Resources

About