Lyme Disease Patient Trajectories Learned from Electronic Medical Data for Stratification of Disease Risk and Therapeutic Response

Ichikawa, Osamu; Glicksberg, Benjamin S.; Genes, Nicholas; Kidd, Brian; Li, Li; Dudley, Joel T.

doi:10.1038/s41598-019-41128-x

Cited by 5 publications

(6 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ichikawa O. et. al [23]. 2019 3639 combinations of samples from patients with Lyme disease and coexisting conditions…”

Section: Elkhadrawi M Et Al [21] 2023mentioning

confidence: 99%

“…decided to investigate it in 2019 when they developed a machine learning algorithm, based on in-depth analysis of electronic medical records, to more accurately describe the diversity of Lyme disease, aiming to create models predicting the identification of drugs associated with the risk of developing additional comorbidities [22,23]. Lyme disease.…”

Section: Elkhadrawi M Et Al [21] 2023mentioning

confidence: 99%

See 1 more Smart Citation

AI and Lyme Disease: Pioneering Advances in Diagnostic Accuracy

Dębiec,

Roman,

Gondko

et al. 2024

J Educ Health Sport

View full text Add to dashboard Cite

Introduction and Purpose: Accurate diagnosis of Lyme disease remains challenging due to its varied manifestations and the limitations of current diagnostic tests. This review examines the emerging role of artificial intelligence (AI) in enhancing the diagnostic accuracy for Lyme disease, aiming to understand how these technologies can be integrated into clinical practice.State of Knowledge: AI and machine learning techniques are increasingly applied to improve diagnostic processes. In Lyme disease, AI models have been developed to identify patterns in clinical data, enhancing early detection and accuracy. Studies have focused on using AI to interpret complex serological results and clinical symptoms more effectively than traditional methods. Additionally, AI has been utilized to analyze geographical and epidemiological data to predict Lyme disease risk areas, aiding in preventive strategies.Summary: AI holds significant promise in transforming Lyme disease diagnostics by increasing the speed and accuracy of detection. These technologies not only help in overcoming the limitations of current serological testing but also provide a framework for predictive analytics in epidemiology. As AI models continue to evolve, their integration into healthcare systems requires careful consideration of ethical implications and validation on broader scales. Future research should focus on refining AI algorithms, improving data inclusivity, and enhancing interoperability with existing medical systems to fully realize AI's potential in battling Lyme disease.

show abstract

“…Ichikawa O. et. al [23]. 2019 3639 combinations of samples from patients with Lyme disease and coexisting conditions…”

Section: Elkhadrawi M Et Al [21] 2023mentioning

confidence: 99%

Section: Elkhadrawi M Et Al [21] 2023mentioning

confidence: 99%

AI and Lyme Disease: Pioneering Advances in Diagnostic Accuracy

Dębiec,

Roman,

Gondko

et al. 2024

J Educ Health Sport

View full text Add to dashboard Cite

show abstract

“…Most cases of this condition resolve following treatment with the antibiotic doxycycline, but more serious cases can lead to an autoimmune response and long-term symptoms [31]. Current informatics research on Lyme disease focuses on identifying clinical factors that can affect treatment as well as on characterizing symptoms more likely to be present after diagnosis [32][33][34]. These methods identify cohorts of patients simply by diagnosis code, due to the lack of a phenotype algorithm.…”

Section: Phe2vec To Extend Disease Phenotype Knowledgementioning

confidence: 99%

“…Lee et al explored the use of embeddings based on GloVe and graph-modeling as well obtaining similar results in terms of phenotype overlap with PheKB [45]. While they included more diseases (33) in the experiments, they did not evaluate the phenotypes in retrieving patients, which is the primary motivation for disease phenotyping. Moreover, they modeled global co-occurrences of medical concepts across the data, without considering the longitudinality of clinical data.…”

Section: Background and Significancementioning

confidence: 99%

Phe2vec: Automated Disease Phenotyping based on Unsupervised Embeddings from Electronic Health Records

Freitas

Johnson

Golden

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

ObjectiveWe introduce Phe2vec, an automated framework for disease phenotyping from electronic health records (EHRs) based on unsupervised learning. We assess its effectiveness against standard rule-based algorithms from the Phenotype KnowledgeBase (PheKB).Materials and MethodsPhe2vec is based on pre-computing embeddings of medical concepts and patients’ longitudinal clinical history. Disease phenotypes are then derived from a seed concept and its neighbors in the embedding space. Patients are similarly linked to a disease if their embedded representation is close to the phenotype. We evaluated Phe2vec using 49,234 medical concepts from structured EHRs and clinical notes from 1,908,741 patients in the Mount Sinai Health System. We assessed performance on ten diverse diseases having a PheKB algorithm, and one disease without, namely Lyme disease.ResultsPhe2vec phenotypes derived using Word2vec, GloVe, and Fasttext embeddings led to promising performance in disease definition and patient cohort identification as compared with standard PheKB definitions. When comparing head-to-head Phe2vec and PheKB disease patient cohorts using chart review, Phe2vec performed on par or better in nine out of ten diseases in terms of predictive positive values. Additionally, Phe2vec effectively identified phenotype definition and patient cohort for Lyme disease, a condition not covered in PheKB.DiscussionPhe2vec offers a solution to improve time-consuming phenotyping pipelines. Differently from other automated approaches in the literature, it is fully unsupervised, can easily scale to any disease and was validated against widely adopted expert-based standards.ConclusionPhe2vec aims to optimize clinical informatics research by augmenting current frameworks to characterize patients by condition and derive reliable disease cohorts.

show abstract

“…Exclusion of Lyme disease was necessary for the differential diagnosis process because the treatment options available would have been contraindicated in some diagnoses. For example, corticosteroids are a treatment option in recurrent fever syndromes but has been reported to increase Lyme disease severity [9,10].…”

Section: Differential Diagnosis and Clinical Coursementioning

confidence: 99%

Cyclical Fevers in a 4-year-old Boy with IgA Deficiency

Yang

2019

Reports

View full text Add to dashboard Cite

While immunodeficiencies, such as (Immunoglobulin A) IgA deficiency, may predispose pediatric patients to respiratory illnesses, they are also associated with autoinflammatory conditions. Distinguishing between these possible complications requires an awareness of these uncommon conditions. We report a case of a four-year-old boy with a history of IgA deficiency presenting to his primary care provider with fever of one week’s duration accompanied by abdominal, ear, and throat pain. The patient’s mother reported that he had these symptoms twice a month for the prior ten months during which he occasionally developed oral ulcers. Between episodes, the patient fully recovered without complications. Considering the patient’s history of IgA deficiency, it was originally assumed that the patient had strep throat and coincidental resolution of his symptoms after initiation of antibiotic therapy supported this diagnosis. However, due to the recurrent nature of his symptoms, a preliminary diagnosis of periodic fever, aphthous stomatitis, pharyngitis, and adenitis (PFAPA) was made. A referral to an otolaryngologist was made, and findings of adenopathy and tonsillitis suggestive of PFAPA were confirmed. The patient was scheduled for adenoidectomy and tonsillectomy, which were performed without complication, leading to the cessation of these episodes.

show abstract

Lyme Disease Patient Trajectories Learned from Electronic Medical Data for Stratification of Disease Risk and Therapeutic Response

Cited by 5 publications

References 42 publications

AI and Lyme Disease: Pioneering Advances in Diagnostic Accuracy

AI and Lyme Disease: Pioneering Advances in Diagnostic Accuracy

Phe2vec: Automated Disease Phenotyping based on Unsupervised Embeddings from Electronic Health Records

Cyclical Fevers in a 4-year-old Boy with IgA Deficiency

Contact Info

Product

Resources

About