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

Internal Medicine Internal Medicine

Topic modeling to characterize the natural history of ANCA-Associated vasculitis from clinical notes: A proof of concept study.

In Seminars in arthritis and rheumatism ; h5-index 49.0

OBJECTIVES : Clinical notes from electronic health records (EHR) are important to characterize the natural history, comorbidities, and complications of ANCA-associated vasculitis (AAV) because these details may not be captured by claims and structured data. However, labor-intensive chart review is often required to extract information from notes. We hypothesized that machine learning can automatically discover clinically-relevant themes across longitudinal notes to study AAV.

METHODS : This retrospective study included prevalent PR3- or MPO-ANCA+ AAV cases managed within the Mass General Brigham integrated health care system with providers' notes available between March 1, 1990 and August 23, 2018. We generated clinically-relevant topics mentioned in notes using latent Dirichlet allocation-based topic modeling and conducted trend analyses of those topics over the 2 years prior to and 5 years after the initiation of AAV-specific treatment.

RESULTS : The study cohort included 660 patients with AAV. We generated 90 topics using 113,048 available notes. Topics were related to the AAV diagnosis, treatment, symptoms and manifestations (e.g., glomerulonephritis), and complications (e.g., end-stage renal disease, infection). AAV-related symptoms and psychiatric symptoms were mentioned months before treatment initiation. Topics related to pulmonary and renal diseases, diabetes, and infections were common during the disease course but followed distinct temporal patterns.

CONCLUSIONS : Automated topic modeling can be used to discover clinically-relevant themes and temporal patterns related to the diagnosis, treatment, comorbidities, and complications of AAV from EHR notes. Future research might compare the temporal patterns in a non-AAV cohort and leverage clinical notes to identify possible AAV cases prospectively.

Wang Liqin, Miloslavsky Eli, Stone John H, Choi Hyon K, Zhou Li, Wallace Zachary S


ANCA-Associated Vasculitis, Electronic Health Records, Epidemiology, Natural Language Processing, Topic Modeling

General General

5D superresolution imaging for a live cell nucleus.

In Current opinion in genetics & development

With a spatial resolution breaking the diffraction limit of light, superresolution imaging allows the visualization of detailed structures of organelles such as mitochondria, cytoskeleton, nucleus, and so on. With multi-dimensional imaging (x, y, z, t, λ), namely, multi-color 3D live imaging enables us fully understand the function of the cell. It is necessary to analyze structural changes or molecular interactions across a large volume in 3D with different labelled targets. To achieve this goal, scientists recently have expanded the original 2D superresolution microscopic tools into 3D imaging techniques. In this review, we will discuss recent development in superresolution microscopy for live imaging with minimal phototoxicity. We will focus our discussion on the cell nucleus where the genetic materials are stored and processed. Machine learning algorism will be introduced to improve the axial resolution of superresolution imaging.

Chu Li-An, Chang Shu-Wei, Tang Wei-Chun, Tseng Yu-Ting, Chen Peilin, Chen Bi-Chang


General General

Design of novel orotansmucosal vaccine-delivery platforms using artificial intelligence.

In European journal of pharmaceutics and biopharmaceutics : official journal of Arbeitsgemeinschaft fur Pharmazeutische Verfahrenstechnik e.V

The linings of the oral cavity are excellent needle-free vaccination sites, able to induce immune responses at distal sites and confer systemic protection. However, owing to the mucosal tissues' intrinsic characteristics, the design of effective antigen-delivery systems is not an easy task. In the present work, we propose to develop and characterize thermosensitive and mucoadhesive hydrogels for orotransmucosal vaccination taking advantage of artificial intelligence tools (AIT). Hydrogels of variable composition were obtained combining Pluronic® F127 (PF127), Hybrane® S1200 (HS1200) and Gantrez® AN119 (AN119) or S97 (S97). Systems were characterized in terms of physicochemical properties, adhesion capacity to mucosal tissues and antigen-like microspheres release. Additionally, polymers biocompatibility and their immune-stimulation capacity was assessed in human macrophages. Interestingly, cells treated with HS1200 exhibited a significant proliferation enhancement compared to control. The use of AIT allowed to determine the effect of each polymer on formulations properties. The used proportions of PF127 and Gantrez® are mainly the factors controlling gelation temperature, mucoadhesion, adhesion work and gel strength. Meanwhile, cohesion and short-term microsphere release are dependent on the PF127 concentration. However, long-term microsphere release varies depending on the Gantrez® variety and the PF127 concentration used. Hydrogels prepared with S97 showed slower microsphere release. The use of AIT allowed to establish the conditions able to produce ternary hydrogels with immune-stimulatory properties together with adequate mucoadhesion capacity and antigen-like microspheres release.

Garcia-Del Rio Lorena, Diaz-Rodriguez Patricia, Landin Mariana


Antigen delivery, Artificial intelligence, Hydrogel design, Mucoadhesive networks, Mucosal vaccination, Thermosensitive platforms

Cardiology Cardiology

A Head-to Head Comparison of Machine Learning Algorithms for Identification of Implanted Cardiac Devices.

In The American journal of cardiology ; h5-index 64.0

Application of artificial intelligence techniques in medicine has rapidly expanded in recent years. Two algorithms for identification of cardiac implantable electronic devices using chest radiography were recently developed: The PacemakerID algorithm, available as a mobile phone application (PIDa) and a web platform (PIDw) and The Pacemaker Identification with Neural Networks (PPMnn), available via web platform. In this study, we assessed the relative accuracy of these algorithms. The machine learning algorithms (PIDa, PIDw, PPMnn) were used to predict device manufacturer using chest X-rays for patients with implanted devices. Each prediction was considered correct if predicted certainty was >75%. For comparative purposes, accuracy of each prediction was compared to the result using the CARDIA-X algorithm. 500 X-rays were included from a convenience sample. Raw accuracy was PIDa 89%, PIDw 73%, PPMnn 71% and CARDIA-X 85%. In conclusion, machine learning algorithms for identification of cardiac devices are accurate at determining device manufacturer, have capacity for improved accuracy with additional training sets and can utilize simple user interfaces. These algorithms have clinical utility in limiting potential infectious exposures and facilitate rapid identification of devices as needed for device reprogramming.

Chudow Jay J, Jones Davis, Weinreich Michael, Zaremski Lynn, Lee Suegene, Weinreich Brian, Krumerman Andrew, Fisher John D, Ferrick Kevin J


General General

Resolving heterogeneity in transcranial electrical stimulation efficacy for attention deficit hyperactivity disorder.

In Experimental neurology

While the treatment of Attention Deficit Hyperactivity Disorder (ADHD) is dominated by pharmacological agents, transcranial electrical stimulation (tES) is gaining attention as an alternative method for treatment. Most current meta-analyses have suggested that tES can improve cognitive functions that are otherwise impaired in ADHD, such as inhibition and working memory, as well as alleviated clinical symptoms. Here we review some of the promising findings in the field of tES. At the same time, we highlight two factors, which hinder the effective application of tES in treating ADHD: 1) the heterogeneity of tES protocols used in different studies; 2) patient profiles influencing responses to tES. We highlight potential solutions for overcoming such limitations, including the use of active machine learning, and provide simulated data to demonstrate how these solutions could also improve the understanding, diagnosis, and treatment of ADHD.

Lipka Renée, Ahlers Eike, Reed Thomas L, Karstens Malin I, Nguyen Vu, Bajbouj Malek, Cohen Kadosh Roi


ADHD, Bayesian optimization, Heterogeneity, NIBS, Personalization, tES

General General

Application of machine learning in the diagnosis of gastric cancer based on noninvasive characteristics.

In PloS one ; h5-index 176.0

BACKGROUND : The diagnosis of gastric cancer mainly relies on endoscopy, which is invasive and costly. The aim of this study is to develop a predictive model for the diagnosis of gastric cancer based on noninvasive characteristics.

AIMS : To construct a predictive model for the diagnosis of gastric cancer with high accuracy based on noninvasive characteristics.

METHODS : A retrospective study of 709 patients at Zhejiang Provincial People's Hospital was conducted. Variables of age, gender, blood cell count, liver function, kidney function, blood lipids, tumor markers and pathological results were analyzed. We used gradient boosting decision tree (GBDT), a type of machine learning method, to construct a predictive model for the diagnosis of gastric cancer and evaluate the accuracy of the model.

RESULTS : Of the 709 patients, 398 were diagnosed with gastric cancer; 311 were health people or diagnosed with benign gastric disease. Multivariate analysis showed that gender, age, neutrophil lymphocyte ratio, hemoglobin, albumin, carcinoembryonic antigen (CEA), carbohydrate antigen 125 (CA125) and carbohydrate antigen 199 (CA199) were independent characteristics associated with gastric cancer. We constructed a predictive model using GBDT, and the area under the receiver operating characteristic curve (AUC) of the model was 91%. For the test dataset, sensitivity was 87.0% and specificity 84.1% at the optimal threshold value of 0.56. The overall accuracy was 83.0%. Positive and negative predictive values were 83.0% and 87.8%, respectively.

CONCLUSION : We construct a predictive model to diagnose gastric cancer with high sensitivity and specificity. The model is noninvasive and may reduce the medical cost.

Zhu Shuang-Li, Dong Jie, Zhang Chenjing, Huang Yao-Bo, Pan Wensheng