Predicting chronic kidney disease progression with artificial intelligence.

BACKGROUND: The use of tools that allow estimation of the probability of progression of chronic kidney disease (CKD) to advanced stages has not yet achieved significant practical importance in clinical setting. This study aimed to develop and validate a machine learning-based model for predicting the need for renal replacement therapy (RRT) and disease progression for patients with stage 3-5 CKD. METHODS: This was a retrospective, closed cohort, observational study. Patients with CKD affiliated with a private insurer with five-year follow-up data were selected. Demographic, clinical, and laboratory variables were included, and the models were developed based on machine learning methods. The outcomes were CKD progression, a significant decrease in the estimated glomerular filtration rate (eGFR), and the need for RRT. RESULTS: Three prediction models were developed-Model 1 (risk at 4.5 years, n = 1446) with a F1 of 0.82, 0.53, and 0.55 for RRT, stage progression, and reduction in the eGFR, respectively,- Model 2 (time- to-event, n = 2143) with a C-index of 0.89, 0.67, and 0.67 for RRT, stage progression, reduction in the eGFR, respectively, and Model 3 (reduced Model 2) with C-index = 0.68, 0.68 and 0.88, for RRT, stage progression, reduction in the eGFR, respectively. CONCLUSION: The time-to-event model performed well in predicting the three outcomes of CKD progression at five years. This model can be useful for predicting the onset and time of occurrence of the outcomes of interest in the population with established CKD.

Structural Protein Effects Underpinning Cognitive Developmental Delay of the PURA p.Phe233del Mutation Modelled by Artificial Intelligence and the Hybrid Quantum Mechanics-Molecular Mechanics Framework.

A whole-exome capture and next-generation sequencing was applied to an 11 y/o patient with a clinical history of congenital hypotonia, generalized motor and cognitive neurodevelopmental delay, and severe cognitive deficit, and without any identifiable Syndromic pattern, and to her parents, we disclosed a de novo heterozygous pathogenic mutation, c.697_699del p.Phe233del (rs786204835)(ACMG classification PS2, PM1, PM2, PP5), harbored in the PURA gene (MIM*600473) (5q31.3), associated with Autosomal Dominant Mental Retardation 31 (MIM # 616158). We used the significant improvement in the accuracy of protein structure prediction recently implemented in AlphaFold that incorporates novel neural network architectures and training procedures based on the evolutionary, physical, and geometric constraints of protein structures. The wild-type (WT) sequence and the mutated sequence, missing the Phe233, were reconstructed. The predicted local Distance Difference Test (lDDT) for the PURAwt and the PURA-Phe233del showed that the occurrence of the Phe233del affects between 220-320 amino acids. The distortion in the PURA structural conformation in the ~5 Å surrounding area after the p.Phe233del produces a conspicuous disruption of the repeat III, where the DNA and RNA helix unwinding capability occurs. PURA Protein-DNA docking corroborated these results in an in silico analysis that showed a loss of the contact of the PURA-Phe233del III repeat domain model with the DNA. Together, (i) the energetic and stereochemical, (ii) the hydropathic indexes and polarity surfaces, and (iii) the hybrid Quantum Mechanics-Molecular Mechanics (QM-MM) analyses of the PURA molecular models demarcate, at the atomic resolution, the specific surrounding region affected by these mutations and pave the way for future cell-based functional analysis. To the best of our knowledge, this is the first report of a de novo mutation underpinning a PURA syndrome in a Latin American patient and highlights the importance of predicting the molecular effects in protein structure using artificial intelligence algorithms and molecular and atomic resolution stereochemical analyses.

A Comprehensive Machine Learning Framework for the Exact Prediction of the Age of Onset in Familial and Sporadic Alzheimer's Disease.

Machine learning (ML) algorithms are widely used to develop predictive frameworks. Accurate prediction of Alzheimer's disease (AD) age of onset (ADAOO) is crucial to investigate potential treatments, follow-up, and therapeutic interventions. Although genetic and non-genetic factors affecting ADAOO were elucidated by other research groups and ours, the comprehensive and sequential application of ML to provide an exact estimation of the actual ADAOO, instead of a high-confidence-interval ADAOO that may fall, remains to be explored. Here, we assessed the performance of ML algorithms for predicting ADAOO using two AD cohorts with early-onset familial AD and with late-onset sporadic AD, combining genetic and demographic variables. Performance of ML algorithms was assessed using the root mean squared error (RMSE), the R-squared (R2), and the mean absolute error (MAE) with a 10-fold cross-validation procedure. For predicting ADAOO in familial AD, boosting-based ML algorithms performed the best. In the sporadic cohort, boosting-based ML algorithms performed best in the training data set, while regularization methods best performed for unseen data. ML algorithms represent a feasible alternative to accurately predict ADAOO with little human intervention. Future studies may include predicting the speed of cognitive decline in our cohorts using ML.

Determinants of adherence to continuous positive airway pressure therapy in adults with obstructive sleep apnea / Determinantes de adherencia a presión positiva continua de la via aérea en adultos con apnea del sueño

ASBTRACT Objectives Adherence to continuous positive airway pressure (CPAP) devices in patients with obstructive sleep apnea (OSA) determines the effectiveness of the treat-ment. Likewise, the assessment of the control of the disease must consider the information referred by the patient, among other value-based health measures related to the satisfaction of the intervention. The objectives of this study were a) Determine the factors related to adherence to CPAP devices in subjects with OSA affiliated to an insurance company of the healthcare system in Colombia. b) Assess symptom control associated to the disease from the individual's perspective and his/her satisfaction with the treatment received. Materials and Methods 1,501 subjects with OSA were surveyed by telephone to explore: sociodemographic factors, habits and lifestyles, use of CPAP and its adverse events, control of the disease, comorbidities, access to care and therapy satisfaction. Using multilevel logistic regression techniques, the influence of the various factors on adherence to CPAP was analyzed, using Stata 13 software. Results Adherence to CPAP therapy was of 58% and the control of symptoms was of 41.7%. The factors that determined the use of CPAP were knowledge on how the device operates, and the disturbances during sleep due to the mask or nasal pad. The-rapy satisfaction was predominantly very good or good. Conclusion Even with moderate adherence values and a good experience with CPAP therapy, symptomatic control of the disease is poor. Many of the factors that affect the use of CPAP are modifiable with a proper approach by the devices' service provider.

RESUMEN Objetivos La adherencia a los dispositivos de presión positiva continua de la vía aérea (CPAP) en pacientes con síndrome de apnea obstructiva del sueño (SAOS) define la efectividad del tratamiento. Los objetivos de este estudio fueron: a) Determinar los factores relacionados con la adherencia al CPAP en pacientes con SAOS de una aseguradora del Sistema General de Seguridad Social en Salud colombiano y b) evaluar el control de los síntomas de la enfermedad desde la perspectiva del individuo y su satisfacción con la terapia. Materiales y Métodos Mediante encuesta telefónica a 1 501 pacientes con SAOS se exploraron factores sociodemográficos, hábitos y estilos de vida, uso del CPAP y eventos adversos relacionados, control de la enfermedad, comorbilidad, acceso a la atención y satisfacción con la terapia. Utilizando técnicas de regresión logística mul-tinivel, se analizó la influencia de los distintos factores sobre la adherencia al CPAP mediante el software Stata 13. Resultados La adherencia al CPAP fue del 58% y el control de los síntomas del 41,7%. Los factores que determinaron el uso del CPAP fueron el conocimiento del funcionamiento del equipo y la dificultad para dormir debida a la mascarilla o la almohadilla nasal. La satisfacción con la terapia fue buena o muy buena predominantemente. Conclusiones Aún con valores de adherencia moderados y una buena experiencia con la terapia CPAP, el control sintomático de la enfermedad es pobre, pues varios de los factores afectan el uso del CPAP. Dichos factores se pueden intervenir con un adecuado abordaje por parte del prestador de servicios del dispositivo.

The cost of hospital care for HIV patients in Colombia: an insurer's perspective.

The aim of this study was to evaluate the cost derived from the hospitalization of people living with HIV (PLHIV) in Colombia between 2011 and 2015. This is an analysis of the direct cost of PLHIV hospitalization from the perspective of an insurer of the Colombian General Social Security System. The costs were calculated in Colombian pesos and corrected for inflation on the basis of the 2017 Consumer Price Index of the Bank of the Republic of Colombia. It was converted to US dollars at the Market Representative Exchange Rate of the same year. We analyzed 1129 hospitalizations in 612 PLHIV, of which 12% started with a diagnosis of HIV during the same hospitalization, with the majority in the AIDS stage (63%). The median overall cost of hospitalizations was US$1509 (25th and 75th percentiles: US$711-US$3254), being even higher in patients with AIDS and as the CD4 T lymphocyte count decreased. The cost derived from the medical care of PLHIV increases as the clinical control of the disease worsens, and it is a key indicator of the impact of the strategies implemented for the timely identification of the infection and subsequent management of the disease.

Targeting Neuroplasticity, Cardiovascular, and Cognitive-Associated Genomic Variants in Familial Alzheimer's Disease.

The identification of novel genetic variants contributing to the widespread in the age of onset (AOO) of Alzheimer's disease (AD) could aid in the prognosis and/or development of new therapeutic strategies focused on early interventions. We recruited 78 individuals with AD from the Paisa genetic isolate in Antioquia, Colombia. These individuals belong to the world largest multigenerational and extended pedigree segregating AD as a consequence of a dominant fully penetrant mutation in the PSEN1 gene and exhibit an AOO ranging from the early 1930s to the late 1970s. To shed light on the genetic underpinning that could explain the large spread of the age of onset (AOO) of AD, 64 single nucleotide polymorphisms (SNP) associated with neuroanatomical, cardiovascular, and cognitive measures in AD were genotyped. Standard quality control and filtering procedures were applied, and single- and multi-locus linear mixed-effects models were used to identify AOO-associated SNPs. A full two-locus interaction model was fitted to define how identified SNPs interact to modulate AOO. We identified two key epistatic interactions between the APOE*E2 allele and SNPs ASTN2-rs7852878 and SNTG1-rs16914781 that delay AOO by up to ~ 8 years (95% CI 3.2-12.7, P = 1.83 × 10-3) and ~ 7.6 years (95% CI 3.3-11.8, P = 8.69 × 10-4), respectively, and validated our previous finding indicating that APOE*E2 delays AOO of AD in PSEN1 E280 mutation carriers. This new evidence involving APOE*E2 as an AOO delayer could be used for developing precision medicine approaches and predictive genomics models to potentially determine AOO in individuals genetically predisposed to AD.

Hemoglobinopathy detection through an institutional neonatal screening program in Colombia / Detecção de hemoglobinopatias por meio de um programa institucional de triagem neonatal na Colômbia

ABSTRACT Introduction: Hemoglobinopathies are among the most common genetic disorders of hemoglobin worldwide and a public health problem. In Colombia, even though geographical areas with high incidence of this disorder have been reported, the absence of a national screening program does not permit us to determine its prevalence. Objective: Establish the prevalence of hemoglobin variants in a population covered by the neonatal screening program of Clínica Colsanitas S.A., between June 2000 and December 2014, including eight capital cities in Colombia. Methods: A retrospective cross-sectional study was conducted. We collected data from reports of the neonatal hemoglobinopathy-screening program for full-term newborn babies between 5 and 15 days old. Qualitative hemoglobin analysis was performed using gel electrophoresis of blood samples taken from the babies' heels. Results: The overall prevalence of abnormal Hb was 1.3%. Within the groups of newborns affected with any hemoglobinopathy (n = 400), the most frequent abnormal structural hemoglobins found were HbS (43%), HbC (9%), fast Hb (8%). For quantitative hemoglobins, HbA2 was 3.7% and HbA kept slightly elevated in 14.7% of cases. Frequency of homozygosis for HbS was 0.01%. Barranquilla, Cartagena and Cali were the cities with the greatest frequency of hemoglobinopathies. No correlation between sex and abnormal hemoglobin was found. Discussion and conclusion: Taking in consideration data from the World Health Organization (WHO) on hemoglobinopathies, our prevalence of > 1% is considered high. Therefore, a more extended coverage and the need for a national screening program are priorities.

RESUMO Introdução: As hemoglobinopatias são doenças genéticas comuns em todo o mundo e representam um problema de saúde pública. Na Colômbia, embora existam áreas geográficas com maior risco de apresentá-las, não há programas de triagem nem estudos para estabelecer sua prevalência na população. Objetivo: Estabelecer a prevalência de variantes de hemoglobina (Hb) na população pertencente ao programa de triagem neonatal da Clínica Colsanitas S.A. entre junho de 2000 e dezembro de 2014, em oito cidades do país. Métodos: Estudo transversal retrospectivo. Os registros do programa de triagem neonatal das hemoglobinopatias foram revistos para a informação dos resultados de eletroforese de hemoglobina em pH alcalino, praticada no sangue dos recém-nascidos com idades compreendidas entre 5-15 dias. Resultados: A prevalência geral de Hb anormal foi de 1,3%. Dentro dos grupos de recém-nascidos afetados com qualquer hemoglobinopatia (n = 400), as hemoglobinas anormais estruturais mais frequentes foram hemoglobina S (HbS) (43%), hemoglobina C (HbC) (9%) e Hb rápida (8%). Para as Hb quantitativas, o aumento da hemoglobina A2 (HbA2) foi de 3,7%, e a hemoglobina A (HbA) aumentada permaneceu ligeiramente elevada em 14,7% casos. A frequência de homozigotos para HbS foi de 0,01%. Barranquilla, Cartagena e Cali foram as cidades com maior frequência de hemoglobinopatias. Não houve associação entre sexo e presença de algum tipo de Hb. Discussão e conclusão: A prevalência global de hemoglobinopatias em nosso estudo foi alta (> 1%) de acordo com os critérios da Organização Mundial de Saúde (OMS). Portanto, há a necessidade de implementação de programas de triagem neonatal com maior cobertura nacional para as hemoglobinopatias.

Autoverification of the automated blood cell counter (CBC) in a reference laboratory in Bogota, Colombia / Autoverificação do contador de células automatizado de sangue (CBC) em um laboratório de referência em Bogotá, Colômbia

ABSTRACT Introduction: The clinical laboratory is part of the group of actors in health systems that are under increasing pressure by users and administrators to increase their productivity in order to respond efficiently to the increased volume of patients, optimizing costs and professional time. This pressure forced laboratories to perform a full review of their procedures and develop technical, logistical and computational tools to enable excellent response times. Objective: This study aimed to evaluate the implementation of the automated blood cell counter autoverification process and its impact on the safety of patients. Methods: Verification rules were designed in the connectivity software, based on manual validation criteria for laboratory professionals, according to the guidelines of the Clinical and Laboratory Standards Institute (CLSI) Guideline Auto10-A and the International Consensus Group for Hematology Review (ISLH). The autoverification percentage was established, and non-conforming product (NCP) percentages were estimated before and after the procedure. Pilot tests were also performed in different days so as to adjust the process. Results: 53.4% of automated blood cell counters autoverification were achieved, and, subsequently in the audit of 18 months, 60% was reached due to verification adjustments in the delta programmed filter. The NCPs rose from 0.065% to 0.0036% from the beginning to the end of the process. Conclusion: The autoverification process enabled to reduce the variability associated with human intervention, therefore the professional is able to focus on the pathological report analysis, reducing the risk of errors and advocating greater importance on patient safety.

RESUMO Introdução: O laboratório clínico é parte do grupo de atores nos sistemas de saúde, que são cada vez mais pressionados por usuários e administradores para aumentar sua produtividade a fim de responder de forma eficiente ao aumento do volume de pacientes, otimizando custos e tempo dos funcionários. Essa pressão forçou laboratórios para efetuar uma revisão completa de seus procedimentos, bem como desenvolver instrumentos técnicos, logísticos e computacionais para permitir excelentes tempos de resposta. Objetivo: Neste estudo, a implementação do processo automatizado de autoverificação do hemograma e seu impacto sobre a segurança do paciente são avaliados. Métodos: Regras de verificação foram construídas no software de conectividade com base em critérios de validação manual dos profissionais de laboratório, de acordo com as orientações do Clinical and Laboratory Standards Institute (CLSI) de Guideline Auto10-A e do International Consensus Group for Hematology Review (ISLH). A percentagem de autoensaio foi estabelecida e as de produtos não conformes (PNC), estimadas antes e depois do procedimento. Testes piloto foram realizados em dias diferentes para ajustar o processo. Resultados: Cinquenta e três por cento da autoverificação dos hemogramas automatizados foram alcançados e, posteriormente, na auditoria dos 18 meses, foram obtidos 60% devido a ajustes na verificação do filtro delta programado. Os PNCs aumentaram de 0,065% a 0,0036% desde o início até ao final do processo. Conclusão: O processo do autoverificação ajudou a reduzir a variabilidade associada à intervenção humana, portanto o profissional pode concentrar-se na análise de relatórios patológicos, reduzindo o risco de erros e dando maior importância para a segurança do paciente.