Predicting Colonic Neoplasia Surgical Complications: A Machine Learning Approach.

BACKGROUND: A range of statistical approaches have been used to help predict outcomes associated with colectomy. The multifactorial nature of complications suggests that machine learning algorithms may be more accurate in determining postoperative outcomes by detecting nonlinear associations, which are not readily measured by traditional statistics. OBJECTIVE: The aim of this study was to investigate the utility of machine learning algorithms to predict complications in patients undergoing colectomy for colonic neoplasia. DESIGN: Retrospective analysis using decision tree, random forest, and artificial neural network classifiers to predict postoperative outcomes. SETTINGS: National Inpatient Sample database (2003-2017). PATIENTS: Adult patients who underwent elective colectomy with anastomosis for neoplasia. MAIN OUTCOME MEASURES: Performance was quantified using sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve to predict the incidence of anastomotic leak, prolonged length of stay, and inpatient mortality. RESULTS: A total of 14,935 patients (4731 laparoscopic, 10,204 open) were included. They had an average age of 67 ± 12.2 years, and 53% of patients were women. The 3 machine learning models successfully identified patients who developed the measured complications. Although differences between model performances were largely insignificant, the neural network scored highest for most outcomes: predicting anastomotic leak, area under the receiver operating characteristic curve 0.88/0.93 (open/laparoscopic, 95% CI, 0.73-0.92/0.80-0.96); prolonged length of stay, area under the receiver operating characteristic curve 0.84/0.88 (open/laparoscopic, 95% CI, 0.82-0.85/0.85-0.91); and inpatient mortality, area under the receiver operating characteristic curve 0.90/0.92 (open/laparoscopic, 95% CI, 0.85-0.96/0.86-0.98). LIMITATIONS: The patients from the National Inpatient Sample database may not be an accurate sample of the population of all patients undergoing colectomy for colonic neoplasia and does not account for specific institutional and patient factors. CONCLUSIONS: Machine learning predicted postoperative complications in patients with colonic neoplasia undergoing colectomy with good performance. Although validation using external data and optimization of data quality will be required, these machine learning tools show great promise in assisting surgeons with risk-stratification of perioperative care to improve postoperative outcomes. See Video Abstract . PREDICCIN DE LAS COMPLICACIONES QUIRRGICAS DE LA NEOPLASIA DE COLON UN ENFOQUE DE MODELO DE APRENDIZAJE AUTOMTICO: ANTECEDENTES:Se han utilizado una variedad de enfoques estadísticos para ayudar a predecir los resultados asociados con la colectomía. La naturaleza multifactorial de las complicaciones sugiere que los algoritmos de aprendizaje automático pueden ser más precisos en determinar los resultados posoperatorios al detectar asociaciones no lineales, que generalmente no se miden en las estadísticas tradicionales.OBJETIVO:El objetivo de este estudio fue investigar la utilidad de los algoritmos de aprendizaje automático para predecir complicaciones en pacientes sometidos a colectomía por neoplasia de colon.DISEÑO:Análisis retrospectivo utilizando clasificadores de árboles de decisión, bosques aleatorios y redes neuronales artificiales para predecir los resultados posoperatorios.AJUSTE:Base de datos de la Muestra Nacional de Pacientes Hospitalizados (2003-2017).PACIENTES:Pacientes adultos sometidos a colectomía electiva con anastomosis por neoplasia.INTERVENCIONES:N/A.PRINCIPALES MEDIDAS DE RESULTADO:El rendimiento se cuantificó utilizando la sensibilidad, especificidad, precisión y la característica operativa del receptor del área bajo la curva para predecir la incidencia de fuga anastomótica, duración prolongada de la estancia hospitalaria y mortalidad de los pacientes hospitalizados.RESULTADOS:Se incluyeron un total de 14.935 pacientes (4.731 laparoscópicos, 10.204 abiertos). Presentaron una edad promedio de 67 ± 12,2 años y el 53% eran mujeres. Los tres modelos de aprendizaje automático identificaron con éxito a los pacientes que desarrollaron las complicaciones medidas. Aunque las diferencias entre el rendimiento del modelo fueron en gran medida insignificantes, la red neuronal obtuvo la puntuación más alta para la mayoría de los resultados: predicción de fuga anastomótica, característica operativa del receptor del área bajo la curva 0,88/0,93 (abierta/laparoscópica, IC del 95%: 0,73-0,92/0,80-0,96); duración prolongada de la estancia hospitalaria, característica operativa del receptor del área bajo la curva 0,84/0,88 (abierta/laparoscópica, IC del 95%: 0,82-0,85/0,85-0,91); y mortalidad de pacientes hospitalizados, característica operativa del receptor del área bajo la curva 0,90/0,92 (abierto/laparoscópico, IC del 95%: 0,85-0,96/0,86-0,98).LIMITACIONES:Los pacientes de la base de datos de la Muestra Nacional de Pacientes Hospitalizados pueden no ser una muestra precisa de la población de todos los pacientes sometidos a colectomía por neoplasia de colon y no tienen en cuenta factores institucionales y específicos del paciente.CONCLUSIONES:El aprendizaje automático predijo con buen rendimiento las complicaciones postoperatorias en pacientes con neoplasia de colon sometidos a colectomía. Aunque será necesaria la validación mediante datos externos y la optimización de la calidad de los datos, estas herramientas de aprendizaje automático son muy prometedoras para ayudar a los cirujanos con la estratificación de riesgos de la atención perioperatoria para mejorar los resultados posoperatorios. (Traducción-Dr. Fidel Ruiz Healy ).

Methods Open - A New Journal Track Starting in 2017.

From 2017 (volume 56) onwards the journal Methods of Information in Medicine will consist of two tracks. Authors can decide to submit their manuscript to either the subscription track that continues to publish its six print and electronic (non-open access) issues for journal subscribers, or the new Methods Open track that will consist of digitally published manuscripts (as gold open access). These two tracks will constitute from 2017 on the journal's Tandem Model. Simultaneously, Methods will introduce a double-blinded review process and reviewer assessment by the submitting authors. Implications of these changes for both authors and reviewers are discussed. With these steps, Methods aims to improve the visibility of the journal and contribute to sharing research results as timely and as widely as possible and thereby to promote scientific progress.

Discussion of "The New Role of Biomedical Informatics in the Age of Digital Medicine".

This article is part of a For-Discussion-Section of Methods of Information in Medicine about the paper "The New Role of Biomedical Informatics in the Age of Digital Medicine" written by Fernando J. Martin-Sanchez and Guillermo H. Lopez-Campos [1]. It is introduced by an editorial. This article contains the combined commentaries invited to independently comment on the paper of Martin-Sanchez and Lopez-Campos. In subsequent issues the discussion can continue through letters to the editor.

Identifying Complementary and Alternative Medicine Usage Information from Internet Resources. A Systematic Review.

OBJECTIVES: Identify and highlight research issues and methods used in studying Complementary and Alternative Medicine (CAM) information needs, access, and exchange over the Internet. METHODS: A literature search was conducted using Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines from PubMed to identify articles that have studied Internet use in the CAM context. Additional searches were conducted at Nature.com and Google Scholar. RESULTS: The Internet provides a major medium for attaining CAM information and can also serve as an avenue for conducting CAM related surveys. Based on the literature analyzed in this review, there seems to be significant interest in developing methodologies for identifying CAM treatments, including the analysis of search query data and social media platform discussions. Several studies have also underscored the challenges in developing approaches for identifying the reliability of CAM-related information on the Internet, which may not be supported with reliable sources. The overall findings of this review suggest that there are opportunities for developing approaches for making available accurate information and developing ways to restrict the spread and sale of potentially harmful CAM products and information. CONCLUSIONS: Advances in Internet research are yet to be used in context of understanding CAM prevalence and perspectives. Such approaches may provide valuable insights into the current trends and needs in context of CAM use and spread.

The mega-matrix tree of life: using genome-scale horizontal gene transfer and sequence evolution data as information about the vertical history of life.

Because horizontal gene transfer can confound the recovery of the largely prokaryotic tree of life (ToL), most genome-based techniques seek to eliminate horizontal signal from ToL analyses, commonly by sieving out incongruent genes and data. This approach greatly limits the number of gene families analysed to a subset thought to be representative of vertical evolutionary history. However, formalized tests have not been performed to determine whether combining the massive amounts of information available in fully sequenced genomes can recover a reasonable ToL. Consequently, we used empirically defined gene homology definitions from a previous study that delineate xenologous gene families (gene families derived from a common transfer event) to generate a massively concatenated, combined-data ToL matrix derived from 323 404 translated open reading frames arranged into 12 381 gene homologue groups coded as amino acid data and 63 336, 64 105, 65 153, 66 922 and 67 109 gene homologue groups coded as gene presence/absence data for 166 fully sequenced genomes. This whole-genome gene presence/absence and amino acid sequence ToL data matrix is composed of 4867 184 characters (a combined data-type mega-matrix). Phylogenetic analysis of this mega-matrix yielded a fully resolved ToL that classifies all three commonly accepted domains of life as monophyletic and groups most taxa in traditionally recognized locations with high support. Most importantly, these results corroborate the existence of a common evolutionary history for these taxa present in both data types that is evident only when these data are analysed in combination. © The Willi Hennig Society 2010.

ORFcurator: molecular curation of genes and gene clusters in prokaryotic organisms.

UNLABELLED: The ability to detect clusters of functionally related genes in multiple microbial genomes has enormous potential for enhancing studies on gene function and microbial evolution. The staggering amount of new genome sequence data presents a largely untapped resource for gene cluster discovery. To date, gene cluster analysis has not been fully automated, and one must rely on manual, tedious and time-consuming manipulation of sequences. To facilitate accurate and rapid identification of conserved gene clusters, we developed a database-driven web application, called ORFcurator. We used ORFcurator to find clusters containing any genes similar to those of the 14-gene Widespread Colonization Island of Actinobacillus actinomycetemcomitans. From 126 genomes, ORFcurator identified all 73 clusters previously determined by manual searching. AVAILABILITY: ORFcurator and all associated scripts are freely available as supplementary information. SUPPLEMENTARY INFORMATION: http://www.genomecurator.org/ORFcurator/

Discovering protein similarity using natural language processing.

Extracting protein interaction relationships from textual repositories, such as MEDLINE, may prove useful in generating novel biological hypotheses. Using abstracts relevant to two known functionally related proteins, we modified an existing natural language processing tool to extract protein interaction terms. We were able to obtain functional information about two proteins, Amyloid Precursor Protein and Prion Protein, that have been implicated in the etiology of Alzheimer's Disease and Creutzfeldt-Jakob Disease, respectively.