Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

General General

Claims-Based Algorithms for Identifying Patients With Pulmonary Hypertension: A Comparison of Decision Rules and Machine-Learning Approaches.

In Journal of the American Heart Association ; h5-index 70.0

Background Real-world healthcare data are an important resource for epidemiologic research. However, accurate identification of patient cohorts-a crucial first step underpinning the validity of research results-remains a challenge. We developed and evaluated claims-based case ascertainment algorithms for pulmonary hypertension (PH), comparing conventional decision rules with state-of-the-art machine-learning approaches. Methods and Results We analyzed an electronic health record-Medicare linked database from two large academic tertiary care hospitals (years 2007-2013). Electronic health record charts were reviewed to form a gold standard cohort of patients with (n=386) and without PH (n=164). Using health encounter data captured in Medicare claims (including patients' demographics, diagnoses, medications, and procedures), we developed and compared 2 approaches for identifying patients with PH: decision rules and machine-learning algorithms using penalized lasso regression, random forest, and gradient boosting machine. The most optimal rule-based algorithm-having ≥3 PH-related healthcare encounters and having undergone right heart catheterization-attained an area under the receiver operating characteristic curve of 0.64 (sensitivity, 0.75; specificity, 0.48). All 3 machine-learning algorithms outperformed the most optimal rule-based algorithm (P<0.001). A model derived from the random forest algorithm achieved an area under the receiver operating characteristic curve of 0.88 (sensitivity, 0.87; specificity, 0.70), and gradient boosting machine achieved comparable results (area under the receiver operating characteristic curve, 0.85; sensitivity, 0.87; specificity, 0.70). Penalized lasso regression achieved an area under the receiver operating characteristic curve of 0.73 (sensitivity, 0.70; specificity, 0.68). Conclusions Research-grade case identification algorithms for PH can be derived and rigorously validated using machine-learning algorithms. Simple decision rules commonly applied in published literature performed poorly; more complex rule-based algorithms may potentially address the limitation of this approach. PH research using claims data would be considerably strengthened through the use of validated algorithms for cohort ascertainment.

Ong Mei-Sing, Klann Jeffrey G, Lin Kueiyu Joshua, Maron Bradley A, Murphy Shawn N, Natter Marc D, Mandl Kenneth D


computable phenotype, machine learning, pulmonary hypertension

General General

Penalized Least Squares for Structural Equation Modeling with Ordinal Responses.

In Multivariate behavioral research

Statistical modeling with sparsity has become an active research topic in the fields of statistics and machine learning. Because the true sparsity pattern of a model is generally unknown aforehand, it is often explored by a sparse estimation procedure, like least absolute shrinkage and selection operator (lasso). In this study, a penalized least squares (PLS) method for structural equation modeling (SEM) with ordinal data is developed. PLS describes data generation by an underlying response approach, and uses a least squares (LS) fitting function to construct a penalized estimation criterion. A numerical simulation was used to compare PLS with existing penalized likelihood (PL) in terms of averaged mean square error, absolute bias, and the correctness of the model. Based on these empirical findings, a hybrid PLS was also proposed to improve both PL and PLS. The hybrid PLS first chooses an optimal sparsity pattern by PL, then estimates model parameters by an unpenalized LS under the model selected by PL. We also extended PLS to cases of mixed type data and multi-group analysis. All proposed methods could be realized in the R package lslx.

Huang Po-Hsien


Structural equation modeling, factor analysis, lasso, penalized least squares, polychoric correlation

Surgery Surgery

Artificial intelligence in cardiothoracic surgery.

In Minerva cardioangiologica

The tremendous and rapid technological advances that humans have achieved in the last decade have definitely impacted how surgical tasks are performed in the operating room (OR). As a high-tech work environment, the contemporary OR has incorporated novel computational systems into the clinical workflow, aiming to optimize processes and support the surgical team. Artificial intelligence (AI) is increasingly important for surgical decision making to help address diverse sources of information, such as patient risk factors, anatomy, disease natural history, patient values and cost, and assist surgeons and patients to make better predictions regarding the consequences of surgical decisions. In this review, we discuss the current initiatives that are using AI in cardiothoracic surgery and surgical care in general. We also address the future of AI and how high-tech ORs will leverage human-machine teaming to optimize performance and enhance patient safety.

Dias Roger D, Shah Julie A, Zenati Marco A


Public Health Public Health

Factors Affecting the Incidence of Hospitalized Pneumonia after Influenza Infection in Korea Using the National Health Insurance Research Database, 2014-2018: Focusing on the Effect of Antiviral Therapy in the 2017 Flu Season.

In Journal of Korean medical science

BACKGROUND : This study aimed to investigate the effect of antiviral therapy following influenza outpatient episodes on the incidence of hospitalized pneumonia episodes, one of secondary complications of influenza.

METHODS : In the National Health Insurance Research Database, data from July 2013 to June 2018 were used. All of the claim data with diagnoses of influenza and pneumonia were converted to episodes of care after applying 100 days of window period. With the 100-day episodes of care, the characteristics of influenza outpatient episodes and antiviral therapy for influenza, the incidence of hospitalized pneumonia episodes following influenza, and the effect of antiviral therapy for influenza on hospitalized pneumonia episodes were investigated.

RESULTS : The crude incidence rate of hospitalized pneumonia after influenza infection was 0.57% in both males and females. Factors affecting hospitalized pneumonia included age, income level except self-employed highest (only in females), municipality, medical institution type, precedent chronic diseases except hepatitis (only in females) and antiviral therapy. In the 2017 flu season, the relative risk was 0.38 (95% confidence interval [CI], 0.29-0.50) in males aged 0-9 and 0.43 (95% CI, 0.32-0.57) in females aged 0-9 without chronic diseases, and it was 0.51 (95% CI, 0.42-0.61) in males aged 0-9 and 0.42 (95% CI, 0.35-0.50) in females aged 0-9 with one or more chronic diseases in the aspect of the effect of antiviral therapy on pneumonia. It suggests that antiviral therapy may decrease the incidence of pneumonia after influenza infection.

CONCLUSION : After outpatient episode incidence of influenza, antiviral treatment has been shown to reduce the incidence of hospitalized pneumonia, especially in infants and children, during pandemic season 2017. Antiviral therapy for influenza is recommended to minimize burden caused by influenza virus infection and to reduce pneumonia. In addition, medical costs of hospitalization may decrease by antiviral therapy, especially in infants and children.

Byeon Kyeong Hyang, Kim Jaiyong, Choi Bo Youl, Kim Jin Yong, Lee Nakyoung


Antiviral Treatment, Episode of Care, Influenza, Pneumonia

Radiology Radiology

Accelerating T2 mapping of the brain by integrating deep learning priors with low-rank and sparse modeling.

In Magnetic resonance in medicine ; h5-index 66.0

PURPOSE : To accelerate T2 mapping with highly sparse sampling by integrating deep learning image priors with low-rank and sparse modeling.

METHODS : The proposed method achieves high-speed T2 mapping by highly sparsely sampling (k, TE)-space. Image reconstruction from the undersampled data was done by exploiting the low-rank structure and sparsity in the T2 -weighted image sequence and image priors learned from training data. The image priors for a single TE were generated from the public Human Connectome Project data using a tissue-based deep learning method; the image priors were then transferred to other TEs using a generalized series-based method. With these image priors, the proposed reconstruction method used a low-rank model and a sparse model to capture subject-dependent novel features.

RESULTS : The proposed method was evaluated using experimental data obtained from both healthy subjects and tumor patients using a turbo spin-echo sequence. High-quality T2 maps at the resolution of 0.9 × 0.9 × 3.0 mm3 were obtained successfully from highly undersampled data with an acceleration factor of 8. Compared with the existing compressed sensing-based methods, the proposed method produced significantly reduced reconstruction errors. Compared with the deep learning-based methods, the proposed method recovered novel features better.

CONCLUSION : This paper demonstrates the feasibility of learning T2 -weighted image priors for multiple TEs using tissue-based deep learning and generalized series-based learning. A new method was proposed to effectively integrate these image priors with low-rank and sparse modeling to reconstruct high-quality images from highly undersampled data. The proposed method will supplement other acquisition-based methods to achieve high-speed T2 mapping.

Meng Ziyu, Guo Rong, Li Yudu, Guan Yue, Wang Tianyao, Zhao Yibo, Sutton Brad, Li Yao, Liang Zhi-Pei


T2 mapping, deep learning, low-rank modeling, quantitative imaging, sparse modeling

oncology Oncology

A Generative Adversarial Network-Based (GAN-Based) Architecture for Automatic Fiducial Marker Detection in Prostate MRI-Only Radiotherapy Simulation Images.

In Medical physics ; h5-index 59.0

PURPOSE : Clinical sites utilizing MRI-only simulation imaging for prostate radiotherapy planning typically use fiducial markers for pretreatment patient positioning and alignment. Fiducial markers appear as small signal voids in MRI images and are often difficult to discern. Existing clinical methods for fiducial marker localization require multiple MRI sequences and/or manual interaction and specialized expertise. In this study, we develop a robust method for automatic fiducial marker detection in MRI simulation images of the prostate and quantify the prostate organ localization accuracy using automatically detected fiducial markers in MRI for pretreatment alignment using cone beam CT (CBCT) images.

METHODS AND MATERIALS : In this study, a deep learning-based algorithm was used to convert MRI images into labelled fiducial marker volumes. 77 prostate cancer patients who received marker implantation prior to MRI and CT simulation imaging were selected for this study. Multiple-Echo T1 -VIBE MRI images were acquired, and images were stratified (at the patient level) based on the presence of intraprostatic calcifications. Ground truth (GT) contours were defined by an expert on MRI using CT images. Training was done using the pix2pix generative adversarial network (GAN) image-to-image translation package and model testing was done using five-fold cross validation. For performance comparison, an experienced medical dosimetrist and a medical physicist each manually contoured fiducial markers in MRI images. The percent of correct detections and F1 classification scores are reported for markers detected using the automatic detection algorithm and human observers. The patient positioning errors were quantified by calculating the target registration errors (TREs) from fiducial marker driven rigid registration between MRI and CBCT images. TREs were quantified for fiducial marker contours defined on MRI by the automatic detection algorithm and the two expert human observers.

RESULTS : 96% of implanted fiducial markers were correctly identified using the automatic detection algorithm. Two expert raters correctly identified 97% and 96% of fiducial markers, respectively. The F1 classification score was 0.68, 0.75 and 0.72 for the automatic detection algorithm and two human raters, respectively. The main source of false discoveries was intraprostatic calcifications. The mean TRE differences between alignments from automatic detection algorithm and human detected markers and GT were less than 1 mm.

CONCLUSIONS : We have developed a deep learning-based approach to automatically detect fiducial markers in MRI-only simulation images in a clinically representative patient cohort. The automatic detection algorithm-predicted markers can allow for patient setup with similar accuracy to independent human observers.

Singhrao Kamal, Fu Jie, Parikh Neil R, Mikaeilian Argin G, Ruan Dan, Kishan Amar U, Lewis John H


Deep Learning, Fiducial Markers, MRI in treatment planning, MRI-Only Simulation