Performance and clinical utility of a new supervised machine-learning pipeline in detecting rare ciliopathy patients based on deep phenotyping from electronic health records and semantic similarity
Abstract Background Rare diseases affect approximately 400 million people worldwide. Many of them suffer from delayed diagnosis. Among them, NPHP1-related renal ciliopathies need to be diagnosed as early as possible as potential treatments have been recently investigated with promising results. Our...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2024-02-01
|
| Series: | Orphanet Journal of Rare Diseases |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s13023-024-03063-7 |
| _version_ | 1827326238217207808 |
|---|---|
| author | Carole Faviez Marc Vincent Nicolas Garcelon Olivia Boyer Bertrand Knebelmann Laurence Heidet Sophie Saunier Xiaoyi Chen Anita Burgun |
| author_facet | Carole Faviez Marc Vincent Nicolas Garcelon Olivia Boyer Bertrand Knebelmann Laurence Heidet Sophie Saunier Xiaoyi Chen Anita Burgun |
| author_sort | Carole Faviez |
| collection | DOAJ |
| description | Abstract Background Rare diseases affect approximately 400 million people worldwide. Many of them suffer from delayed diagnosis. Among them, NPHP1-related renal ciliopathies need to be diagnosed as early as possible as potential treatments have been recently investigated with promising results. Our objective was to develop a supervised machine learning pipeline for the detection of NPHP1 ciliopathy patients from a large number of nephrology patients using electronic health records (EHRs). Methods and results We designed a pipeline combining a phenotyping module re-using unstructured EHR data, a semantic similarity module to address the phenotype dependence, a feature selection step to deal with high dimensionality, an undersampling step to address the class imbalance, and a classification step with multiple train-test split for the small number of rare cases. The pipeline was applied to thirty NPHP1 patients and 7231 controls and achieved good performances (sensitivity 86% with specificity 90%). A qualitative review of the EHRs of 40 misclassified controls showed that 25% had phenotypes belonging to the ciliopathy spectrum, which demonstrates the ability of our system to detect patients with similar conditions. Conclusions Our pipeline reached very encouraging performance scores for pre-diagnosing ciliopathy patients. The identified patients could then undergo genetic testing. The same data-driven approach can be adapted to other rare diseases facing underdiagnosis challenges. |
| first_indexed | 2024-03-07T14:41:38Z |
| format | Article |
| id | doaj.art-554f62835a944de6bb50e39eb453ba23 |
| institution | Directory Open Access Journal |
| issn | 1750-1172 |
| language | English |
| last_indexed | 2024-03-07T14:41:38Z |
| publishDate | 2024-02-01 |
| publisher | BMC |
| record_format | Article |
| series | Orphanet Journal of Rare Diseases |
| spelling | doaj.art-554f62835a944de6bb50e39eb453ba232024-03-05T20:20:07ZengBMCOrphanet Journal of Rare Diseases1750-11722024-02-0119111210.1186/s13023-024-03063-7Performance and clinical utility of a new supervised machine-learning pipeline in detecting rare ciliopathy patients based on deep phenotyping from electronic health records and semantic similarityCarole Faviez0Marc Vincent1Nicolas Garcelon2Olivia Boyer3Bertrand Knebelmann4Laurence Heidet5Sophie Saunier6Xiaoyi Chen7Anita Burgun8Centre de Recherche des Cordeliers, Université Paris Cité, Sorbonne Université, INSERM UMR 1138Université Paris Cité, Imagine Institute, Data Science Platform, INSERM UMR 1163Centre de Recherche des Cordeliers, Université Paris Cité, Sorbonne Université, INSERM UMR 1138Department of Pediatric Nephrology, APHP-Centre, Reference Center for Inherited Renal Diseases (MARHEA), Imagine Institute, Hôpital Necker-Enfants Malades, Université Paris CitéNephrology and Transplantation Department, MARHEA, Hôpital Necker-Enfants Malades, AP-HP, Université Paris CitéDepartment of Pediatric Nephrology, APHP-Centre, Reference Center for Inherited Renal Diseases (MARHEA), Imagine Institute, Hôpital Necker-Enfants Malades, Université Paris CitéLaboratory of Renal Hereditary Diseases, INSERM UMR 1163, Imagine Institute, Université Paris CitéCentre de Recherche des Cordeliers, Université Paris Cité, Sorbonne Université, INSERM UMR 1138Centre de Recherche des Cordeliers, Université Paris Cité, Sorbonne Université, INSERM UMR 1138Abstract Background Rare diseases affect approximately 400 million people worldwide. Many of them suffer from delayed diagnosis. Among them, NPHP1-related renal ciliopathies need to be diagnosed as early as possible as potential treatments have been recently investigated with promising results. Our objective was to develop a supervised machine learning pipeline for the detection of NPHP1 ciliopathy patients from a large number of nephrology patients using electronic health records (EHRs). Methods and results We designed a pipeline combining a phenotyping module re-using unstructured EHR data, a semantic similarity module to address the phenotype dependence, a feature selection step to deal with high dimensionality, an undersampling step to address the class imbalance, and a classification step with multiple train-test split for the small number of rare cases. The pipeline was applied to thirty NPHP1 patients and 7231 controls and achieved good performances (sensitivity 86% with specificity 90%). A qualitative review of the EHRs of 40 misclassified controls showed that 25% had phenotypes belonging to the ciliopathy spectrum, which demonstrates the ability of our system to detect patients with similar conditions. Conclusions Our pipeline reached very encouraging performance scores for pre-diagnosing ciliopathy patients. The identified patients could then undergo genetic testing. The same data-driven approach can be adapted to other rare diseases facing underdiagnosis challenges.https://doi.org/10.1186/s13023-024-03063-7Diagnosis supportElectronic health recordSupervised machine learningSemantic similarityImbalanced datasetRare disease |
| spellingShingle | Carole Faviez Marc Vincent Nicolas Garcelon Olivia Boyer Bertrand Knebelmann Laurence Heidet Sophie Saunier Xiaoyi Chen Anita Burgun Performance and clinical utility of a new supervised machine-learning pipeline in detecting rare ciliopathy patients based on deep phenotyping from electronic health records and semantic similarity Orphanet Journal of Rare Diseases Diagnosis support Electronic health record Supervised machine learning Semantic similarity Imbalanced dataset Rare disease |
| title | Performance and clinical utility of a new supervised machine-learning pipeline in detecting rare ciliopathy patients based on deep phenotyping from electronic health records and semantic similarity |
| title_full | Performance and clinical utility of a new supervised machine-learning pipeline in detecting rare ciliopathy patients based on deep phenotyping from electronic health records and semantic similarity |
| title_fullStr | Performance and clinical utility of a new supervised machine-learning pipeline in detecting rare ciliopathy patients based on deep phenotyping from electronic health records and semantic similarity |
| title_full_unstemmed | Performance and clinical utility of a new supervised machine-learning pipeline in detecting rare ciliopathy patients based on deep phenotyping from electronic health records and semantic similarity |
| title_short | Performance and clinical utility of a new supervised machine-learning pipeline in detecting rare ciliopathy patients based on deep phenotyping from electronic health records and semantic similarity |
| title_sort | performance and clinical utility of a new supervised machine learning pipeline in detecting rare ciliopathy patients based on deep phenotyping from electronic health records and semantic similarity |
| topic | Diagnosis support Electronic health record Supervised machine learning Semantic similarity Imbalanced dataset Rare disease |
| url | https://doi.org/10.1186/s13023-024-03063-7 |
| work_keys_str_mv | AT carolefaviez performanceandclinicalutilityofanewsupervisedmachinelearningpipelineindetectingrareciliopathypatientsbasedondeepphenotypingfromelectronichealthrecordsandsemanticsimilarity AT marcvincent performanceandclinicalutilityofanewsupervisedmachinelearningpipelineindetectingrareciliopathypatientsbasedondeepphenotypingfromelectronichealthrecordsandsemanticsimilarity AT nicolasgarcelon performanceandclinicalutilityofanewsupervisedmachinelearningpipelineindetectingrareciliopathypatientsbasedondeepphenotypingfromelectronichealthrecordsandsemanticsimilarity AT oliviaboyer performanceandclinicalutilityofanewsupervisedmachinelearningpipelineindetectingrareciliopathypatientsbasedondeepphenotypingfromelectronichealthrecordsandsemanticsimilarity AT bertrandknebelmann performanceandclinicalutilityofanewsupervisedmachinelearningpipelineindetectingrareciliopathypatientsbasedondeepphenotypingfromelectronichealthrecordsandsemanticsimilarity AT laurenceheidet performanceandclinicalutilityofanewsupervisedmachinelearningpipelineindetectingrareciliopathypatientsbasedondeepphenotypingfromelectronichealthrecordsandsemanticsimilarity AT sophiesaunier performanceandclinicalutilityofanewsupervisedmachinelearningpipelineindetectingrareciliopathypatientsbasedondeepphenotypingfromelectronichealthrecordsandsemanticsimilarity AT xiaoyichen performanceandclinicalutilityofanewsupervisedmachinelearningpipelineindetectingrareciliopathypatientsbasedondeepphenotypingfromelectronichealthrecordsandsemanticsimilarity AT anitaburgun performanceandclinicalutilityofanewsupervisedmachinelearningpipelineindetectingrareciliopathypatientsbasedondeepphenotypingfromelectronichealthrecordsandsemanticsimilarity |