Esperanza Pato-Cour
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Esperanza Pato-Cour has a total of 64 co-authors. Here are the top co-authors by number of shared publications:
Aina Moll-Udina
Ophthalmology Department, Hospital Clínic of Barcelona, Barcelona, Spain
3 shared publicationsAlejandro Fonollosa
Ophthalmology Department, BioCruces-Bizkaia Health Research Institute, Cruces University Hospital, University of the Basque Country, Barakaldo, Spain
3 shared publications- Alfredo Madrid-García
Grupo de Patología Musculoesquelética, Hospital Clínico San Carlos, Instituto de Investigación Sanitaria San Carlos (IdISSC), Madrid, Spain
3 shared publications - Benjamín Fernández-Gutiérrez
Rheumatology Department, Hospital Clínico San Carlos, 28040 Madrid, Spain
3 shared publications - Dalifer Freites-Núñez
Grupo de Patología Musculoesquelética. Hospital Clínico San Carlos. Instituto de Investigación Sanitaria San Carlos (IdISSC)
3 shared publications Fayna Rodríguez-González
Ophthalmology Department, Hospital de Gran Canaria Doctor Negrín, Gran Canaria, Spain
3 shared publications
Latest Publications
- Open Access
- 18 April 2024
- Published by Springer Nature in Ophthalmology and Therapy
- Vol. 13 (6) , 1669-1682
Interobserver Reliability and Sensitivity to Change of a Composite Ocular Inflammatory Activity Index: UVEDAI©
Introduction This was a multicenter, prospective, longitudinal, observational study involving eight Spanish tertiary hospitals to determine the interobserver reliability of an uveitis disease activity index, (UVEDAI) and assess its sensitivity to change in patients with receiving pharmacologic treatment. Methods Patients aged ≥ 18 years diagnosed with active noninfectious uveitis were included. A complete baseline assessment was performed by two ophthalmologists who determined ocular inflammatory activity using the UVEDAI index independently of each other. The principal ophthalmologist made a new visit at 4 weeks to determine the change in inflammatory activity. The interobserver reliability analysis was performed by calculating the intraclass correlation coefficient (ICC), with the values of the variables and the UVEDAI obtained by both ophthalmologists in the more active eye at the baseline visit. Sensitivity to change in the UVEDAI index was assessed at 4 weeks from the start of pharmacologic treatment by determining the clinically relevant change, defined as a change in UVEDAI of ≥ 0.8 points over baseline. The mean change between both measures was compared using the repeated-measures t-test. Results A total of 111 patients were included. In the interobserver reliability analysis, the ICC for the UVEDAI value was 0.9, and, when compared with the mean UVEDAI values obtained by the ophthalmologists, no statistically significant differences were found (p value > 0.05). As for the sensitivity to change in UVEDAI, statistically significant differences (p value = 0.00) were found for the mean values of the index compared with baseline. In all cases, the index value decreased by > 1 point at the 4-week visit. Conclusions The interobserver reliability of the UVEDAI was high in the total sample. Furthermore, the index was sensitive in determining the change in inflammatory activity after treatment. We believe that UVEDAI is a disease activity index that enables objective comparison of results in clinical practice and trials.
- 18 April 2024
- Published by Springer Nature in Ophthalmology and Therapy
- Vol. 13 (6) , 1669-1682
Open AccessInterobserver Reliability and Sensitivity to Change of a Composite Ocular Inflammatory Activity Index: UVEDAI©
Introduction This was a multicenter, prospective, longitudinal, observational study involving eight Spanish tertiary hospitals to determine the interobserver reliability of an uveitis disease activity index, (UVEDAI) and assess its sensitivity to change in patients with receiving pharmacologic treatment. Methods Patients aged ≥ 18 years diagnosed with active noninfectious uveitis were included. A complete baseline assessment was performed by two ophthalmologists who determined ocular inflammatory activity using the UVEDAI index independently of each other. The principal ophthalmologist made a new visit at 4 weeks to determine the change in inflammatory activity. The interobserver reliability analysis was performed by calculating the intraclass correlation coefficient (ICC), with the values of the variables and the UVEDAI obtained by both ophthalmologists in the more active eye at the baseline visit. Sensitivity to change in the UVEDAI index was assessed at 4 weeks from the start of pharmacologic treatment by determining the clinically relevant change, defined as a change in UVEDAI of ≥ 0.8 points over baseline. The mean change between both measures was compared using the repeated-measures t-test. Results A total of 111 patients were included. In the interobserver reliability analysis, the ICC for the UVEDAI value was 0.9, and, when compared with the mean UVEDAI values obtained by the ophthalmologists, no statistically significant differences were found (p value > 0.05). As for the sensitivity to change in UVEDAI, statistically significant differences (p value = 0.00) were found for the mean values of the index compared with baseline. In all cases, the index value decreased by > 1 point at the 4-week visit. Conclusions The interobserver reliability of the UVEDAI was high in the total sample. Furthermore, the index was sensitive in determining the change in inflammatory activity after treatment. We believe that UVEDAI is a disease activity index that enables objective comparison of results in clinical practice and trials.
- Open Access
- 13 December 2023
- Published by Springer Nature in Scientific Reports
- Vol. 13 (1) , 1-11
Harnessing ChatGPT and GPT-4 for evaluating the rheumatology questions of the Spanish access exam to specialized medical training
The emergence of large language models (LLM) with remarkable performance such as ChatGPT and GPT-4, has led to an unprecedented uptake in the population. One of their most promising and studied applications concerns education due to their ability to understand and generate human-like text, creating a multitude of opportunities for enhancing educational practices and outcomes. The objective of this study is twofold: to assess the accuracy of ChatGPT/GPT-4 in answering rheumatology questions from the access exam to specialized medical training in Spain (MIR), and to evaluate the medical reasoning followed by these LLM to answer those questions. A dataset, RheumaMIR, of 145 rheumatology-related questions, extracted from the exams held between 2010 and 2023, was created for that purpose, used as a prompt for the LLM, and was publicly distributed. Six rheumatologists with clinical and teaching experience evaluated the clinical reasoning of the chatbots using a 5-point Likert scale and their degree of agreement was analyzed. The association between variables that could influence the models’ accuracy (i.e., year of the exam question, disease addressed, type of question and genre) was studied. ChatGPT demonstrated a high level of performance in both accuracy, 66.43%, and clinical reasoning, median (Q1–Q3), 4.5 (2.33–4.67). However, GPT-4 showed better performance with an accuracy score of 93.71% and a median clinical reasoning value of 4.67 (4.5–4.83). These findings suggest that LLM may serve as valuable tools in rheumatology education, aiding in exam preparation and supplementing traditional teaching methods.
- 13 December 2023
- Published by Springer Nature in Scientific Reports
- Vol. 13 (1) , 1-11
Open AccessHarnessing ChatGPT and GPT-4 for evaluating the rheumatology questions of the Spanish access exam to specialized medical training
The emergence of large language models (LLM) with remarkable performance such as ChatGPT and GPT-4, has led to an unprecedented uptake in the population. One of their most promising and studied applications concerns education due to their ability to understand and generate human-like text, creating a multitude of opportunities for enhancing educational practices and outcomes. The objective of this study is twofold: to assess the accuracy of ChatGPT/GPT-4 in answering rheumatology questions from the access exam to specialized medical training in Spain (MIR), and to evaluate the medical reasoning followed by these LLM to answer those questions. A dataset, RheumaMIR, of 145 rheumatology-related questions, extracted from the exams held between 2010 and 2023, was created for that purpose, used as a prompt for the LLM, and was publicly distributed. Six rheumatologists with clinical and teaching experience evaluated the clinical reasoning of the chatbots using a 5-point Likert scale and their degree of agreement was analyzed. The association between variables that could influence the models’ accuracy (i.e., year of the exam question, disease addressed, type of question and genre) was studied. ChatGPT demonstrated a high level of performance in both accuracy, 66.43%, and clinical reasoning, median (Q1–Q3), 4.5 (2.33–4.67). However, GPT-4 showed better performance with an accuracy score of 93.71% and a median clinical reasoning value of 4.67 (4.5–4.83). These findings suggest that LLM may serve as valuable tools in rheumatology education, aiding in exam preparation and supplementing traditional teaching methods.
- Open Access
Harnessing ChatGPT and GPT-4 for Evaluating the Rheumatology Questions of the Spanish Access Exam to Specialized Medical Training
The emergence of Large Language Models (LLM) with remarkable performance such as ChatGPT and GPT-4, has led to an unprecedented uptake in the population. One of their most promising and studied applications concerns education due to their ability to understand and generate human-like text, creating a multitude of opportunities for enhancing educational practices and outcomes. The objective of this study is two-fold: to assess the accuracy of ChatGPT/GPT-4 in answering rheumatology questions from the access exam to specialized medical training in Spain (MIR), and to evaluate the medical reasoning followed by these LLM to answer those questions. A dataset, RheumaMIR, of 145 rheumatology-related questions, extracted from the exams held between 2010 and 2023, was created for that purpose, used as a prompt for the LLM, and was publicly distributed. Six rheumatologists with clinical and teaching experience evaluated the clinical reasoning of the chatbots using a 5-point Likert scale and their degree of agreement was analyzed. The association between variables that could influence the models’ accuracy (i.e., year of the exam question, disease addressed, type of question and genre) was studied. ChatGPT demonstrated a high level of performance in both accuracy, 66.43%, and clinical reasoning, median (Q1-Q3), 4.5 (2.33-4.67). However, GPT-4 showed better performance with an accuracy score of 93.71% and a median clinical reasoning value of 4.67 (4.5-4.83). These findings suggest that LLM may serve as valuable tools in rheumatology education, aiding in exam preparation and supplementing traditional teaching methods. What is already known on this topic Large Language Models have demonstrated remarkable performance when presented with medical exam questions. However, no study has evaluated their clinical reasoning in the rheumatology field. What this study adds This is the first study to evaluate the accuracy and clinical reasoning of ChatGPT and GPT-4 when rheumatology questions from an official access exam to specialized medical training are used as prompts. How this study might affect research, practice or policy?This study highlights the usefulness of two Large Language Models, ChatGPT and GPT-4, in the training of medical students in the field of rheumatology. Highlights ChatGPT showed an accuracy of 66.43% in answering MIR questions, while GPT-4 exhibits a significantly higher proficiency with an accuracy of 93.71%. The median (Q1-Q3) value of the average score for the clinical reasoning of GPT-4 was 4.67 (4.5-4.83), while for ChatGPT was 4.5 (2.33-4.67).
Open AccessHarnessing ChatGPT and GPT-4 for Evaluating the Rheumatology Questions of the Spanish Access Exam to Specialized Medical Training
The emergence of Large Language Models (LLM) with remarkable performance such as ChatGPT and GPT-4, has led to an unprecedented uptake in the population. One of their most promising and studied applications concerns education due to their ability to understand and generate human-like text, creating a multitude of opportunities for enhancing educational practices and outcomes. The objective of this study is two-fold: to assess the accuracy of ChatGPT/GPT-4 in answering rheumatology questions from the access exam to specialized medical training in Spain (MIR), and to evaluate the medical reasoning followed by these LLM to answer those questions. A dataset, RheumaMIR, of 145 rheumatology-related questions, extracted from the exams held between 2010 and 2023, was created for that purpose, used as a prompt for the LLM, and was publicly distributed. Six rheumatologists with clinical and teaching experience evaluated the clinical reasoning of the chatbots using a 5-point Likert scale and their degree of agreement was analyzed. The association between variables that could influence the models’ accuracy (i.e., year of the exam question, disease addressed, type of question and genre) was studied. ChatGPT demonstrated a high level of performance in both accuracy, 66.43%, and clinical reasoning, median (Q1-Q3), 4.5 (2.33-4.67). However, GPT-4 showed better performance with an accuracy score of 93.71% and a median clinical reasoning value of 4.67 (4.5-4.83). These findings suggest that LLM may serve as valuable tools in rheumatology education, aiding in exam preparation and supplementing traditional teaching methods. What is already known on this topic Large Language Models have demonstrated remarkable performance when presented with medical exam questions. However, no study has evaluated their clinical reasoning in the rheumatology field. What this study adds This is the first study to evaluate the accuracy and clinical reasoning of ChatGPT and GPT-4 when rheumatology questions from an official access exam to specialized medical training are used as prompts. How this study might affect research, practice or policy?This study highlights the usefulness of two Large Language Models, ChatGPT and GPT-4, in the training of medical students in the field of rheumatology. Highlights ChatGPT showed an accuracy of 66.43% in answering MIR questions, while GPT-4 exhibits a significantly higher proficiency with an accuracy of 93.71%. The median (Q1-Q3) value of the average score for the clinical reasoning of GPT-4 was 4.67 (4.5-4.83), while for ChatGPT was 4.5 (2.33-4.67).
- Open Access
- 20 February 2023
- Published by Springer Nature in Ophthalmology and Therapy
- Vol. 12 (2) , 1057-1058
Correction to: Validation of UVEDAI: An Index for Evaluating the Level of Inflammatory Activity in Uveitis
- 20 February 2023
- Published by Springer Nature in Ophthalmology and Therapy
- Vol. 12 (2) , 1057-1058
Open AccessCorrection to: Validation of UVEDAI: An Index for Evaluating the Level of Inflammatory Activity in Uveitis
- Open Access
- 23 January 2023
- Published by Springer Nature in Ophthalmology and Therapy
- Vol. 12 (2) , 1045-1055
Validation of UVEDAI: An Index for Evaluating the Level of Inflammatory Activity in Uveitis
Introduction Uveitis is the inflammation of the middle layer of the eye, the uvea, and is a major cause of blindness. None of the instruments used in clinical practice are, in themselves, sufficient to evaluate the course of uveitis. Therefore, it is necessary to develop instruments enabling standardized measurement of inflammatory activity. We developed a composite disease activity index for patients with uveitis known as UVEDAI, which considers the overall activity of the eye. The objective of this study was to validate the composite index of ocular inflammation, UVEDAI. Methods A multicenter cross-sectional study involving eight Spanish tertiary hospitals. Sixty-two patients aged ≥ 18 years with acute uveitis were recruited. Participants gave informed consent before participating in the study. A full ophthalmological examination was performed by two ophthalmologists to determine inflammatory activity: one used the UVEDAI score and the other used clinical judgment. The ophthalmologists did not share their findings with each other to avoid introducing bias into the analysis. Construct validity was established by means of factor analysis. The criterion validity of the index was determined using an ordinal multivariate regression model, in which the dependent variable was the degree of uveal inflammation (mild, moderate, or high/severe). Cut-off points were determined for the UVEDAI and for the receiver operating characteristic (ROC) curves. Results Sixty-two patients were included. Total variance with the three components accounted for 80.32% of the construct validity. Each of the three components identified one type of eye involvement. The discriminatory capacity of UVEDAI was 0.867 (95% CI 0.778; 0.955 p < 0.001) for mild versus moderate–high and 0.946 (95% CI 0.879; 1.000 p < 0.001) for high versus mild–moderate. Conclusions The variables included in UVEDAI enable ocular inflammatory activity to be described with a high degree of accuracy. The index may be used to evaluate and classify this activity with considerable discriminatory power.
- 23 January 2023
- Published by Springer Nature in Ophthalmology and Therapy
- Vol. 12 (2) , 1045-1055
Open AccessValidation of UVEDAI: An Index for Evaluating the Level of Inflammatory Activity in Uveitis
Introduction Uveitis is the inflammation of the middle layer of the eye, the uvea, and is a major cause of blindness. None of the instruments used in clinical practice are, in themselves, sufficient to evaluate the course of uveitis. Therefore, it is necessary to develop instruments enabling standardized measurement of inflammatory activity. We developed a composite disease activity index for patients with uveitis known as UVEDAI, which considers the overall activity of the eye. The objective of this study was to validate the composite index of ocular inflammation, UVEDAI. Methods A multicenter cross-sectional study involving eight Spanish tertiary hospitals. Sixty-two patients aged ≥ 18 years with acute uveitis were recruited. Participants gave informed consent before participating in the study. A full ophthalmological examination was performed by two ophthalmologists to determine inflammatory activity: one used the UVEDAI score and the other used clinical judgment. The ophthalmologists did not share their findings with each other to avoid introducing bias into the analysis. Construct validity was established by means of factor analysis. The criterion validity of the index was determined using an ordinal multivariate regression model, in which the dependent variable was the degree of uveal inflammation (mild, moderate, or high/severe). Cut-off points were determined for the UVEDAI and for the receiver operating characteristic (ROC) curves. Results Sixty-two patients were included. Total variance with the three components accounted for 80.32% of the construct validity. Each of the three components identified one type of eye involvement. The discriminatory capacity of UVEDAI was 0.867 (95% CI 0.778; 0.955 p < 0.001) for mild versus moderate–high and 0.946 (95% CI 0.879; 1.000 p < 0.001) for high versus mild–moderate. Conclusions The variables included in UVEDAI enable ocular inflammatory activity to be described with a high degree of accuracy. The index may be used to evaluate and classify this activity with considerable discriminatory power.
- Open Access
- 2 August 2019
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 8 (8) , 1156
Outpatient Readmission in Rheumatology: A Machine Learning Predictive Model of Patient’s Return to the Clinic
Our objective is to develop and validate a predictive model based on the random forest algorithm to estimate the readmission risk to an outpatient rheumatology clinic after discharge. We included patients from the Hospital Clínico San Carlos rheumatology outpatient clinic, from 1 April 2007 to 30 November 2016, and followed-up until 30 November 2017. Only readmissions between 2 and 12 months after the discharge were analyzed. Discharge episodes were chronologically split into training, validation, and test datasets. Clinical and demographic variables (diagnoses, treatments, quality of life (QoL), and comorbidities) were used as predictors. Models were developed in the training dataset, using a grid search approach, and performance was compared using the area under the receiver operating characteristic curve (AUC-ROC). A total of 18,662 discharge episodes were analyzed, out of which 2528 (13.5%) were followed by outpatient readmissions. Overall, 38,059 models were developed. AUC-ROC, sensitivity, and specificity of the reduced final model were 0.653, 0.385, and 0.794, respectively. The most important variables were related to follow-up duration, being prescribed with disease-modifying anti-rheumatic drugs and corticosteroids, being diagnosed with chronic polyarthritis, occupation, and QoL. We have developed a predictive model for outpatient readmission in a rheumatology setting. Identification of patients with higher risk can optimize the allocation of healthcare resources.
- 2 August 2019
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 8 (8) , 1156
Open AccessOutpatient Readmission in Rheumatology: A Machine Learning Predictive Model of Patient’s Return to the Clinic
Our objective is to develop and validate a predictive model based on the random forest algorithm to estimate the readmission risk to an outpatient rheumatology clinic after discharge. We included patients from the Hospital Clínico San Carlos rheumatology outpatient clinic, from 1 April 2007 to 30 November 2016, and followed-up until 30 November 2017. Only readmissions between 2 and 12 months after the discharge were analyzed. Discharge episodes were chronologically split into training, validation, and test datasets. Clinical and demographic variables (diagnoses, treatments, quality of life (QoL), and comorbidities) were used as predictors. Models were developed in the training dataset, using a grid search approach, and performance was compared using the area under the receiver operating characteristic curve (AUC-ROC). A total of 18,662 discharge episodes were analyzed, out of which 2528 (13.5%) were followed by outpatient readmissions. Overall, 38,059 models were developed. AUC-ROC, sensitivity, and specificity of the reduced final model were 0.653, 0.385, and 0.794, respectively. The most important variables were related to follow-up duration, being prescribed with disease-modifying anti-rheumatic drugs and corticosteroids, being diagnosed with chronic polyarthritis, occupation, and QoL. We have developed a predictive model for outpatient readmission in a rheumatology setting. Identification of patients with higher risk can optimize the allocation of healthcare resources.
- Open Access
- 4 November 2016
- Published by Springer Nature in Rheumatology International
- Vol. 37 (4) , 647-656
Development of an activity disease score in patients with uveitis (UVEDAI)
- 4 November 2016
- Published by Springer Nature in Rheumatology International
- Vol. 37 (4) , 647-656
Open AccessDevelopment of an activity disease score in patients with uveitis (UVEDAI)