A Machine Learning Tool for Early Identification of Celiac Disease Autoimmunity

doi:10.21203/rs.3.rs-3891354/v1

Download PDF

Article

A Machine Learning Tool for Early Identification of Celiac Disease Autoimmunity

https://doi.org/10.21203/rs.3.rs-3891354/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Identifying which patients should undergo serologic screening for celiac disease (CD) may help diagnose patients who otherwise often experience diagnostic delays or remain undiagnosed. Using anonymized outpatient data from the electronic medical records of Maccabi Healthcare Services, we developed and evaluated a gradient boosted trees model to classify patients as at-risk for CD autoimmunity prior to first documented diagnosis or positive serum tissue transglutaminase (tTG-IgA). A train set of highly seropositive (tTG-IgA > 10X ULN) cases (n = 677) with likely CD and controls (n = 176,293) with no evidence of CD autoimmunity was used for model development. Input features included demographic information and commonly available laboratory results. The model was then evaluated for discriminative ability as measured by AUC on a distinct set of highly seropositive cases (n = 153) and controls (n = 41,087). Performance was estimated at one (AUC = 0.84), two (AUC = 0.81), three (AUC = 0.82) and four (AUC = 0.79) years prior to first documented evidence of disease. The model’s ability to distinguish cases of incident CD autoimmunity from controls shows promise as a potential clinical tool to identify patients at increased risk for having undiagnosed celiac disease in the community for serologic screening.

Health sciences/Gastroenterology/Gastrointestinal diseases

Health sciences/Medical research/Translational research

Health sciences/Medical research/Pre clinical studies

Physical sciences/Mathematics and computing/Information technology

Health sciences/Risk factors

Celiac disease (CD) is an immune-mediated disease characterized by small bowel enteropathy triggered by dietary exposure to gluten. The classic clinical presentation of CD includes signs of malabsorption and gastrointestinal symptoms such as abdominal pain, bloating and diarrhea [1]. However, many patients experience predominantly non-specific extra-intestinal symptomatology [2, 3], or are asymptomatic [4]. The diverse manifestations of CD can make it challenging for primary care providers (PCPs) to identify and diagnose CD in the general population [5]. Indeed, a majority of adult patients with CD today are likely undiagnosed [4, 5, 6, 7]. Patients who are eventually receive a diagnosis experience a mean delay of eleven years from symptom onset to diagnosis, with more than half reporting a delay of five or more years till the diagnosis is established [5]. CD has an estimated global prevalence of over 1% [8], and a rising incidence in recent years [5, 9, 10]. Diagnostic delays and underdiagnosis of CD therefore represent an important healthcare problem.

Screening for CD is typically performed by highly sensitive and widely available serum tests for CD autoimmunity (CDA): the most accurate and commonly used being the assay for antibodies to tissue-transglutaminase (tTG-IgA) [11]. Seropositivity is suggestive of underlying CD, and such patients typically undergo endoscopic evaluation to establish or rule out the diagnosis of CD via biopsy of intestinal mucosa [12]. High seropositivity (tTG-IgA > 10X ULN) is associated with a high (> 95%) positive predictive value (PPV) for villous atrophy and, in the proper clinical setting, is considered sufficient for diagnosis without a biopsy in children and possibly in adults [6, 13, 14, 15].

Clinical guidelines do not provide clear and consistent definitions on when to screen for CDA. General population screening is not currently recommended [16], although screening high-risk patients may be warranted [17, 18]. What defines high-risk groups for CD varies somewhat between reports, although the focus has typically been on a family history of CD and medical comorbidities with established associations with CD [19]. Beyond these factors, laboratory abnormalities such as iron-deficiency anemia are common among patients with CD [20]. While PCPs may be aware of specific risk-factors, signs and symptoms of undiagnosed CD, more subtle combinations of clinical features may go missed.

Machine learning (ML) algorithms have the potential to use existing data within a patient’s electronic medical record (EMR) to provide risk assessments to providers [21]. ML models have been developed to alert intensive care unit physicians to patients at risk of circulatory failure [22], to identify clinically significant portal hypertension in non-alcoholic steatohepatitis patients from pathology reports [23], to predict incident hypertension [24] and hypertension outcomes [25], to identify patients with undiagnosed psoriatic arthritis [26], to predict dementia onset [27], future Parkinson’s disease diagnosis [28], and to flag patients at risk of advanced colorectal cancer [29]. One previous study that attempted to develop ML models to identify patients with incident CD using a variety of modeling methods found that the models were not consistently better than chance [30]. Another study showed positive results, but the study size was small and the models relied on symptoms extracted from unstructured clinical documents and diagnostic codes [31]. Neither study included objective laboratory test results as predictive input features, which may have hindered performance and limited generalizability.

The goal of the current study was to develop and assess a prescreening EMR-based tool to classify adult patients by risk of having unidentified CD autoimmunity using ML methods. Input features included commonly available laboratory results, age and biological sex. Incident cases were identified from a large retrospective community-based dataset using results from tTG-IgA testing. Highly seropositive cases (tTG-IgA > 10X ULN) with probable underlying CD were used for model training and evaluation against cohorts of controls with no evidence of disease. Performance was additionally assessed in a test set consisting of a cohort of seropositive cases (tTG-IgA > 2X ULN) who may require endoscopic evaluation for CD and a cohort of controls. In both test sets discriminative ability as measured by estimated area under the ROC (AUC) was assessed at multiple time points before first documented evidence of CD autoimmunity.

Dataset

The dataset for this retrospective study consisted of deidentified EMR data from Maccabi Health Services (MHS), Israel’s 2nd largest health maintenance organization with over 2.7 million documented members [32]. Data were accessed via the Kahn Sagol Maccabi Research and Innovation Centre (KSM), and extracted using the MDClone platform (version 5.5.0.4). The dataset was deidentified by KSM, and no personal identifying information was made available to the researchers. Approval for use of the dataset and the retrospective analysis was obtained by the Maccabi institutional review board (approval #0052-20-MHS), and the study was conducted in accordance with the Declaration of Helsinki and all relevant guidelines and regulations. Informed consent was waived as all identifying information had been removed by KSM.

Study cohort definitions

Patients were eligible for inclusion if they (1) were MHS members during the study period (2005–2021), (2) were born before 1996, and (3) joined MHS before 2005. The eligible population was randomly split into train (80%) and test (20%) sets. In this study we distinguish between seropositive cases in general and highly seropositive cases with likely underlying CD. Cohorts of highly seropositive cases were selected in both train and test sets, and an additional cohort of seropositive cases was identified using case identification criteria (CIC) described below.

Highly seropositive cases were defined as patients having at least one documented tTG-IgA test ≥ 10X ULN, which has an extremely high (> 95%) PPV for duodenal biopsy proven CD [6, 14, 15, 33]. CIC for seropositive cases included all patients with at least one documented tTG-IgA > 2X ULN. This definition is more sensitive for underlying CD, but also includes a higher proportion of cases that would not have pathologic evidence of disease [6]. Reports of procedures including endoscopic evaluations were not available to researchers, so pathologic evidence of CD could not be confirmed or ruled out.

To reduce the possibility that cases were not newly diagnosed, cases were excluded if they had a history of tTG-IgA levels within normal limits before their first positive tTG-IgA, or a CD diagnosis code > 1 year before their first positive tTG-IgA. Providers may order serology on patients with known CD to check patient compliance with a gluten-free diet (GFD), so such cases may have been previously diagnosed and therefore not incident cases relevant for this study. For each seropositive case, the earlier of the first positive serology or first diagnosis of CD was defined as that patient’s index date.

Screening serology was performed with one of two commercial kits used during the study period: Celiakey (Thermo Fisher Scientific-USA) for years 2005–2011 and Elia (Thermo Fisher Scientific-USA) for years 2012–2021, each with manufacturer established ULN values of > 5 U/mL, and > 7 U/mL respectively.

Controls were identified as eligible MHS patients with no documented CD diagnosis code and no serologic evidence of CD autoimmunity. One cohort of controls was constituted from the train set and two cohorts of controls in the test set were selected to match the two cohorts of cases. Controls were matched to cases by years of data availability at the maximal possible ratio of controls to cases. Controls were assigned the same index dates as the case to which they were matched. Controls were not matched by demographic characteristics to allow the model to learn the relationship between these features for predicting incident CD seropositivity.

Each patient included in the train or test sets was additionally assigned a run date, defined as the first of July of the calendar year preceding the patient’s index date. This gap between run date and index date is referred to as the one-year gap, referring to the mean time between run date and index dates of the cohort. The gap between the run date and the index date was added to account for potential clinical suspicion for CD immediately preceding the index date. Analyses were also conducted at further time gaps of up to four years prior to index dates to test the ability of the model to identify patients years before intial suspicion of disease. This methodology was described in detail in a previous report on patients with psoriatic arthritis [26]. Patients were additionally excluded if on their run date they did not meet the following criteria: (i) age ≥ 18 years-old, (ii) members of MHS for at least four years, and (iii) at least one complete blood count (CBC) during the four years prior to their index date.

Model development

The model was trained using gradient boosted trees (XGBoost), a particularly robust ML method for tabular data, which often produces the best performance compared to other ML methods [25, 34, 35, 36]. The model was trained and evaluated using data available during the three years prior to the run date. Training was performed at a one-year time gap between run dates and index dates.

Candidate independent predictors consisted of basic demographic information (biological sex and age at run date), and laboratory test results extracted from the structured data (Supplementary Table 1). The most recent available laboratory result of each type log-transformed by sex and age group. Missing data were given a special indication and the model treated these as null values. SHapley Additive exPlanations (SHAPs) were used both for model selection and to explain the output of the model [37]. To select only the highest contributing features, the model was retrained with the 20 features with the highest SHAP importance values.

Model Evaluation

To test the ability of the model to identify incident CD seropositivity prior to the first documented evidence of disease, model performance was assessed at run dates at one-, two-, three- and four-year time gaps prior to each patient’s index date. Tests at each time gap were performed on the same base cohort selected according to the criteria described above. Patients from each test cohort who did not meet eligibility criteria at previous time gaps were removed from analyses. Discriminative performance at each time gap was evaluated by AUC. In addition, to test the association between the model’s classification and incident seropositivity, odds ratios (ORs) are reported for the patients with model scores in the top 1%, corresponding to a specificity of approximately 99%. The high specificity threshold was selected to reflect a potential clinical use case.

Cohorts of cases of CDA and controls in the train and test sets are described in Table 1.

Table 1: Descriptive characteristics of the train and test cohorts. The first test cohort includes cases of celiac disease autoimmunity with high seropositivity (tTG-IgA>10X ULN) and matched controls. The second test cohort includes all cases of seropositivity with tTG-IgA>2X ULN.

Train Cohort	Patients, n	Age, mean (SD), y	Female sex, n (%)
Controls	176,293	50.7 (18.2)	110,978 (63.0%)
Highly seropositive cases	677	37.8 (17.1)	499 (73.7%)

Highly seropositive CD autoimmunity	Patients, n	Age, mean (SD), y	Female sex, n (%)
Controls	41,087	49.9 (17.5)	25,657 (62.4%)
Highly seropositive cases	153	36.2 (16.5)	114 (74.5%)

Seropositive CD autoimmunity	Patients, n	Age, mean (SD), y	Female sex, n (%)
Controls	78,923	49.9 (17.5)	49,483 (62.8%)
Seropositive cases	301	37.1 (17.1)	210 (69.8%)

Performance measures on the high seropositivity test cohort at gaps of one, two, three and four years are shown in Fig. 1. At the one-year gap, discriminatory AUC was 0.84. Feature contributions are depicted for this test by SHAP analysis (Fig. 2). At the two-, three-, and four-year gaps, AUC values were 0.81, 0.82 and 0.79 respectively. Estimated ORs for flagging incident cases at the 99% specificity cutoff at the one-, two-, three- and four-year time gaps were: 18.53 [CI: 12.03–28.54], 17.84 [CI: 11.26–28.26], 17.71 [CI: 10.95–28.67], 7.87 [CI: 3.78–16.36].

The model’s ability to distinguish between seropositive cases and controls was assessed by AUC as 0.79, 0.77, 0.76, 0.76 at gaps of one, two, three and four years. Estimated ORs that a flagged patient is a case at 99% specificity were 13.86 [CI: 9.87–19.47], 9.60 [CI: 6.27–14.69], 11.39 [CI: 7.35–17.65], 9.79 [CI: 5.90-16.23] at each time gap respectively.

In the current study, we describe the development and evaluation of a novel ML model for identifying adult patients with incident CD autoimmunity up to four years prior to the first documented evidence of disease. The model showed excellent discriminatory ability (AUC > 0.80) between cases of highly seropositive patients with likely CD at gaps of one, two and three years. The model also showed good discriminatory ability (AUC > 0.7) four years prior to initial evidence of highly seropositive CD autoimmunity, and at all time gaps for more broadly defined seropositive cases compared to controls. ORs that a flagged patient is a case were estimated at a high cutoff of approximately 99% specificity to reflect a potential use case as a clinical decision support tool for PCPs. Flagged patients were much more likely to be cases with incident CD autoimmunity at this threshold than patients not flagged, suggesting potential utility as a prescreening tool to identify patients at risk of having undiagnosed CD or CDA. The final model achieved these results using twenty commonly available laboratory and demographic features.

The relationships between demographic and lab features of the model as depicted in the SHAP are consistent with established phenomena among patients with untreated CD. Decreased ferritin and hemoglobin are characteristic of anemia secondary to malabsorption of dietary iron and chronic inflammation in the setting of CD [1]. Liver function tests (LFTs), specifically alanine transaminase (ALT) and aspartate transferase (AST), are elevated in 20%-50% of patients with newly diagnosed CD, and undetected CD accounts for an estimated 4% of cases of unexplained transaminitis [38, 39]. Low high density lipoprotein (HDL) has consistently been associated with untreated CD [40], and the combination of unexplained iron-deficiency anemia and low HDL may be particularly suggestive of CD [41]. Longitudinal studies have found that these lab abnormalities typically resolve after initiation of a GFD, including anemia [42], transaminitis [43], and low HDL [44, 45].

Earlier identification and treatment of CD have been shown to have clinical benefits: screen-detected asymptomatic and mildly symptomatic adult patients typically show improved intestinal histology and reduced serum autoantibody levels after following adherence to a GFD [46, 47]. Late diagnosis in contrast has been associated with unfavorable clinical outcomes. Patients diagnosed after 40 years of age are more likely than younger patients to show persistent signs of intestinal mucosal injury, including villous blunting and intraepithelial lymphocytes in the duodenum despite following a strict GFD [48]. Among symptomatic patients, diagnostic delays are associated with poorer long-term outcomes even after initiation of a GFD, including persistent gastrointestinal and extra-gastrointestinal symptoms [2, 3, 47], increased utilization of healthcare resources, and lower reported quality of life measures [7, 50, 51]. Nevertheless CD is a highly heterogeneous disease, and the long-term benefits of early identification by screening should be established in future studies [52].

To our knowledge, this is the first report of a ML model showing the ability to identify cases of CDA from controls within a large community-based setting using only commonly available laboratory results, biological sex and age in adults. A major strength of the study was the size and continuity of follow-up in the longitudinal data set, which allowed for large patient cohorts with rich historical data. Follow-up studies are needed to evaluate the robustness of the models to other datasets with different population demographics and EMR systems. Additionally, prospective studies should explore the PPV that a flagged patient is seropositive for CD autoimmunity, and has undiagnosed CD.

The study also had limitations. Due to lack of access to unstructured data such as clinical documentation, patients were selected by the proxy of test results rather than by endoscopy results. Some patients selected as cases may not have had CD, although highly seropositive patients are very likely to have CD [6]. Some controls may also have CD that was not documented or undiagnosed at the time the study was conducted. Additionally, there may be heterogeneity in the full population that was not captured in the cohorts of cases and controls used for model training and evaluation. Future studies should examine the robustness of the model, and its ability to perform on different populations and health care systems that may differ from MHS through retrospective and prospective validation studies.

In conclusion, this study presents a ML model based on a large population dataset that can identify adults at risk of undiagnosed CD autoimmunity using commonly available structured objective and demographic data. This model may be clinically useful as a prescreening tool to enhance diagnosis of CD in the community.

Contributor Statement

Michael Dreyfuss: Conceptualization, Manuscript preparation – original draft. Benjamin Getz: Conceptualisation, Data curation, Formal analysis, Methodology, Project administration, Software, Supervision, Validation, Visualisation, Writing – original draft. Benjamin Lebwohl: Conceptualisation, Supervision, Validation, Writing – original draft, Writing – review & editing. Or Ramni: Conceptualisation, Data curation, Formal analysis, Methodology, Project administration, Software, Validation, Visualisation, Writing – original draft. Daniel Underberger: Conceptualisation, Validation, Visualisation, Writing – review & editing. Tahel Ilan Ber: Conceptualisation, Validation, Visualisation. Shlomit Steinberg-Koch: Conceptualisation, Funding acquisition, Project administration, Visualisation, Writing – review & editing. Yonathan Jenudi: Conceptualisation, Formal analysis, Software, Validation, Writing – review & editing. Sivan Gazit: Writing – review & editing. Tal Patalon: Writing – review & editing. Gabriel Chodick: Writing – review & editing. Prof. (Emer.) Yehuda Shoenfeld: Conceptualisation, Supervision Validation, Writing – review & editing. Amir Ben-Tov: Conceptualisation, Methodology, Project administration, Supervision, Validation, Writing – original draft.

ALL authors approved the final version of the article, including the authorship list

Competing Interest

Authors' declaration of personal interests:

Michael Dreyfuss, Benjamin Getz, Or Ramni, Daniel Underberger, Shlomit Steinberg-Koch, Yonathan Jenudi are paid employees of Predicta Med. Tahel Ilan Ber is a former employee of Predicta Med. Kahn Sagol Maccabi Research and Innovation Centre (KSM) is under a commercial agreement with Predicta Med; Sivan Gazit, Tal Patalon, Gabriel Chodick and Amir Ben-Tov are employed by KSM. The rest have no conflict of interest to declare.

Funding

This study was funded in full by Predicta-Med LTD and the Israel Innovation Authority (IIA).

Data Availability:

The authors do not have permission to share the data itself, which can be accessed through agreement with KSM. Statistics and analytics can be provided upon request from Michael Dreyufss or Benjamin Getz.

Ethics Approval Statement:

Ethical approval was obtained from MHS’s Institutional Review Board (IRB) (Approval #0052-20-MHS).

Reilly, N. R., Fasano, A., & Green, P. H. R. Presentation of celiac disease. Gastrointest. Endosc. Clin. N Am. 22, 613–621, (2012).
Laurikka, P., Nurminen, S., Kivelä, L., & Kurppa, K. Extraintestinal manifestations of celiac disease: Early detection for better long-term outcomes. Nutrients 10, 1015 (2018).
Jericho, H., Sansotta, N., & Guandalini, S. Extraintestinal manifestations of celiac disease: Effectiveness of the gluten-free diet. J. Pediatr. Gastroenterol. Nutr. 65, 75–79 (2017).
Choung, R. S., et al. Prevalence and morbidity of undiagnosed celiac disease from a community-based study. Gastroenterology 152, 830-839.e5 (2017).
Lebwohl, B., & Rubio-Tapia, A. Epidemiology, presentation, and diagnosis of celiac disease. Gastroenterology 160, 63-75 (2021).
Kvamme, J.-M., Sørbye, S., Florholmen, J., & Halstensen, T. S. Population-based screening for celiac disease reveals that the majority of patients are undiagnosed and improve on a gluten-free diet. Sci. Rep. 12, 12647 (2022).
Fuchs, V., et al. Delayed celiac disease diagnosis predisposes to reduced quality of life and incremental use of health care services and medicines: A prospective nationwide study. United European Gastroenterol. J. 6, 567–575 (2018).
Singh, P., et al. Global prevalence of celiac disease: Systematic review and meta-analysis. Clin. Gastroenterol. Hepatol., 16, 823-836.e2 (2018).
King, J. A., et al. Incidence of celiac disease is increasing over time. Am. J. Gastroenterol., 115, 1 (2020).
Ludvigsson, J. F., et al. Increasing incidence of celiac disease in a North American population. Am. J. Gastroenterol., 108, 818–824 (2013).
Pelkowski, T. D., & Viera, A. J. Celiac disease: Diagnosis and management. Am. Family Physician, 89, 99–105 (2014).
Ianiro, G. Endoscopic tools for the diagnosis and evaluation of celiac disease. World J. Gastroenterol., 19, 8562 (2013).
Al-Toma, A., et al. European Society for the Study of Coeliac Disease (ESsCD) guideline for coeliac disease and other gluten-related disorders. United European Gastroenterol. J., 7, 583–613 (2019).
Husby, S., et al. European Society for Pediatric Gastroenterology, Hepatology, and Nutrition Guidelines for the Diagnosis of Coeliac Disease. J. Pediatr. Gastroenterol. Nutr., 54, 136–160 (2012).
Ciacci, C., et al. Serum anti-tissue transglutaminase IgA and prediction of duodenal villous atrophy in adults with suspected coeliac disease without IgA deficiency (Bi.A.CeD): A multicentre, prospective cohort study. Lancet Gastroenterol. Hepatol., 8, 1005–1014 (2023).
Bibbins-Domingo, K., et al. Screening for celiac disease. JAMA, 317, 1252 (2017).
Ludvigsson, J. F., et al. Screening for celiac disease in the general population and in high-risk groups. United European Gastroenterol. J., 3, 106–120 (2014).
Richey, R., Howdle, P., Shaw, E., & Stokes, T. Recognition and assessment of coeliac disease in children and adults: Summary of NICE guidance. BMJ, 338, b1684–b1684 (2009).
Chou, R., et al. Screening for celiac disease. JAMA, 317, 1258 (2017).
Agardh, D., et al. Clinical features of celiac disease: A prospective birth cohort. Pediatrics, 135, 627–634 (2015).
Goldstein, B. A., Navar, A. M., Pencina, M. J., & Ioannidis, J. P. A. Opportunities and challenges in developing risk prediction models with electronic health records data: A systematic review. J. Am. Med. Inform. Assoc., 24, 198–208 (2016).
Hyland, S. L., et al. Early prediction of circulatory failure in the intensive care unit using machine learning. Nat. Med., 26, 364–373 (2020).
Bosch, J., et al. A machine learning approach to liver histological evaluation predicts clinically significant portal hypertension in NASH cirrhosis. Hepatology, 74, 3146–3160 (2021).
Ye, C., et al. Prediction of incident hypertension within the next year: Prospective study using statewide electronic health records and machine learning. J. Med. Internet Res., 20, e22 (2018).
Chang, W., et al. A machine-learning-based prediction method for hypertension outcomes based on medical data. Diagnostics, 9, 178 (2019).
Shapiro, J., et al. Evaluation of a machine learning tool for the early identification of patients with undiagnosed psoriatic arthritis – A retrospective population-based study. J. Transl. Autoimmun., 7, 100207–100207 (2023).
Nori, V. S., et al. Machine learning models to predict onset of dementia: A label learning approach. Alzheimer's Dement., 5, 918–925 (2019).
Yuan, W., et al. Accelerating diagnosis of Parkinson’s disease through risk prediction. BMC Neurol., 21, (2021).
Wang, Y.-H., Nguyen, P. A., Islam, M. M., Li, Y.-C., & Yang, H.-C. Development of deep learning algorithm for detection of colorectal cancer in EHR data. Stud Health Technol Inform., 264, 438–441.
Hujoel, I. A., et al. Machine learning in detection of undiagnosed celiac disease. Clin. Gastroenterol. Hepatol., 16, 1354-1355.e1 (2018).
Ludvigsson, J. F., et al. Use of computerized algorithm to identify individuals in need of testing for celiac disease. J. Am. Med. Inform. Assoc., 20, e306-310 (2013).
Gazit, S., et al. The incidence of SARS-CoV-2 reinfection in persons with naturally acquired immunity with and without subsequent receipt of a single dose of BNT162b2 vaccine. Ann. Intern. Med., 175, 674–681 (2022).
Werkstetter, K. J., et al. Accuracy in diagnosis of celiac disease without biopsies in clinical practice. Gastroenterology, 153, 924–935 (2017).
Piccialli, F., et al. Precision medicine and machine learning towards the prediction of the outcome of potential celiac disease. Sci. Rep., 11, (2021).
Lee, S., Lee, H., Choi, J. R., & Koh, S. B. Development and validation of prediction model for risk reduction of metabolic syndrome by body weight control: A prospective population-based study. Sci. Rep., 10, (2020).
Shwartz-Ziv, R., & Armon, A. Tabular data: Deep learning is not all you need. Inf. Fusion, 81, 84–90 (2022).
Lundberg, S., & Lee, S.-I. A unified approach to interpreting model predictions. arXiv preprint arXiv:1705.07874 (2017).
Sainsbury, A., Sanders, D. S., & Ford, A. C. Meta-analysis: Coeliac disease and hypertransaminasaemia. Aliment. Pharmacol. Ther., 34, 33–40 (2011).
Volta, U. Pathogenesis and clinical significance of liver injury in celiac disease. Clin. Rev. Allergy Immunol., 36, 62–70 (2008).
Salardi, S., et al. Whole lipid profile and not only HDL cholesterol is impaired in children with coexisting type 1 diabetes and untreated celiac disease. Acta Diabetol., 54, 889–894 (2017).
Abu Daya, H., Lebwohl, B., Smukalla, S., Lewis, S. K., & Green, P. H. Utilizing HDL levels to improve detection of celiac disease in patients with iron deficiency anemia. Am. J. Gastroenterol., 109, 769–770 (2014).
Bergamaschi, G., et al. Anemia of chronic disease and defective erythropoietin production in patients with celiac disease. Haematologica, 93, 1785–1791 (2008).
Jena, A., et al. Liver abnormalities in celiac disease and response to gluten-free diet: A systematic review and meta-analysis. J. Gastroenterol. Hepatol., 38, 11–22 (2022).
Brar, P., et al. Change in lipid profile in celiac disease: Beneficial effect of gluten-free diet. Am. J. Med., 119, 786–790 (2006).
Nikniaz, Z., Farhangi, M. A., Hosseinifard, H., & Nikniaz, L. Does a gluten-free diet increase body mass index and lipid profile in celiac patients? A systematic review and meta-analysis. Mediterr. J. Nutr. Metab., 12, 341–352 (2019).
Kurppa, K., et al. Benefits of a gluten-free diet for asymptomatic patients with serologic markers of celiac disease. Gastroenterology, 147, 610-617.e1 (2014).
Vilppula, A., et al. Clinical benefit of gluten-free diet in screen-detected older celiac disease patients. BMC Gastroenterol., 11, (2011).
Patel, N. et al. Clinical data do not reliably predict duodenal histology at follow-up in celiac disease. Am. J. Surg. Pathol., 48, 212–220 (2023).
Galli, G., et al. Relationship between persistent gastrointestinal symptoms and duodenal histological findings after adequate gluten-free diet: A gray area of celiac disease management in adult patients. Nutrients, 13, 600 (2021).
Norström, F., Lindholm, L., Sandström, O., Nordyke, K., & Ivarsson, A. Delay to celiac disease diagnosis and its implications for health-related quality of life. BMC Gastroenterol., 11, (2011).
Mårild, K., et al. Costs and use of health care in patients with celiac disease: A population-based longitudinal study. Am. J. Gastroenterol., 115, 1253–1263 (2020).
Paavola, S., et al. Coeliac disease re‐screening among once seronegative at‐risk relatives: A long‐term follow‐up study. United European Gastroenterol. J., 10, 585–593 (2022).

Competing interest reported. Michael Dreyfuss, Benjamin Getz, Or Ramni, Daniel Underberger, Shlomit Steinberg-Koch, Yonathan Jenudi are paid employees of Predicta Med. Tahel Ilan Ber is a former employee of Predicta Med. Kahn Sagol Maccabi Research and Innovation Centre (KSM) is under a commercial agreement with Predicta Med; Sivan Gazit, Tal Patalon, Gabriel Chodick and Amir Ben-Tov are employed by KSM. The rest have no conflict of interest to declare.

CandidateLaboratoryFeatures.xlsx

Download PDF

Editorial decision: Revision requested
24 Feb, 2024
Editor assigned by journal
19 Feb, 2024
Editor invited by journal
12 Feb, 2024
Submission checks completed at journal
12 Feb, 2024
First submitted to journal
23 Jan, 2024

You are reading this latest preprint version

A Machine Learning Tool for Early Identification of Celiac Disease Autoimmunity

Status:

Version 1

Abstract

Figures

Introduction

Methods

Dataset

Study cohort definitions

Model development

Model Evaluation

Results

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1