Artificial Intelligence in Medicine

Reliability-based robust multi-atlas label fusion for brain MRI segmentation Publication date: May 2019 Source: Artificial Intelligence in Medicine, Volume 96 Author(s): Liang Sun, Chen Zu, Wei Shao, Junye Guang, Daoqiang Zhang, Mingxia Liu Abstract Label fusion is one of the key steps in multi-atlas based segmentation of structural magnetic resonance (MR) images. Although a number of label fusion methods have been developed in literature, most of those existing methods fail to address two important problems, i.e., (1) compared with boundary voxels, inner voxels usually have higher probability (or reliability) to be correctly segmented, and (2) voxels with high segmentation reliability (after initial segmentation) can help refine the segmentation of voxels with low segmentation reliability in the target image. To this end, we propose a general reliability-based robust label fusion framework for multi-atlas based MR image segmentation. Specifically, in the first step, we perform initial segmentation for MR images using a conventional multi-atlas label fusion method. In the second step, for each voxel in the target image, we define two kinds of reliability, including the label reliability and spatial reliability that are estimated based on the soft label and spatial information from the initial segmentation, respectively. Finally, we employ voxels with high label-spatial reliability to help refine the label fusion process of those with low reliability in the target image. We incorporate our proposed framework into four well-known label fusion methods, including locally-weighted voting (LWV), non-local mean patch-based method (PBM), joint label fusion (JLF) and sparse patch-based method (SPBM), and obtain four novel label-spatial reliability-based label fusion approaches (called ls-LWV, ls-PBM, ls-JLF, and ls-SPBM). We validate the proposed methods in segmenting ROIs of brain MR images from the NIREP, LONI-LPBA40 and ADNI datasets. The experimental results demonstrate that our label-spatial reliability-based label fusion methods outperform the state-of-the-art methods in multi-atlas image segmentation.

Fast density-peaks clustering for registration-free pediatric white matter tract analysis Publication date: May 2019 Source: Artificial Intelligence in Medicine, Volume 96 Author(s): Xin Fan, Yuzhuo Duan, Shichao Cheng, Yuxi Zhang, Hua Cheng Abstract Clustering white matter (WM) tracts from diffusion tensor imaging (DTI) is primarily important for quantitative analysis on pediatric brain development. A recently developed algorithm, density peaks (DP) clustering, demonstrates great robustness to the complex structural variations of WM tracts without any prior templates. Nevertheless, the calculation of densities, the core step of DP, is time consuming especially when the number of WM fibers is huge. In this paper, we propose a fast algorithm that accelerates the density computation about 50 times over the original one. We convert the global calculation for the density as well as critical parameter in the process into local computations, and develop a binary tree structure to orderly store the neighbors for these local computations. Hence, the density computation turns out to be a direct access of the structure, rendering significantly computational saving. Performing experiments on synthetic point data and the JHU-DTI data set and comparing results of our fast DP algorithm and existing clustering methods, we can validate the efficiency and effectiveness of our fast DP algorithm. Finally, we demonstrate the application of the proposed algorithm on the analysis of pediatric WM tract development.

Retinal blood vessel extraction employing effective image features and combination of supervised and unsupervised machine learning methods Publication date: April 2019 Source: Artificial Intelligence in Medicine, Volume 95 Author(s): Mahdi Hashemzadeh, Baharak Adlpour Azar Abstract In medicine, retinal vessel analysis of fundus images is a prominent task for the screening and diagnosis of various ophthalmological and cardiovascular diseases. In this research, a method is proposed for extracting the retinal blood vessels employing a set of effective image features and combination of supervised and unsupervised machine learning techniques. Further to the common features used in extracting blood vessels, three strong features having a significant influence on the accuracy of the vessel extraction are utilized. The selected combination of the different types of individually efficient features results in a rich local information with better discrimination for vessel and non-vessel pixels. The proposed method first extracts the thick and clear vessels in an unsupervised manner, and then, it extracts the thin vessels in a supervised way. The goal of the combination of the supervised and unsupervised methods is to deal with the problem of intra-class high variance of image features calculated from various vessel pixels. The proposed method is evaluated on three publicly available databases DRIVE, STARE and CHASE_DB1. The obtained results (DRIVE: Acc = 0.9531, AUC = 0.9752; STARE: Acc = 0.9691, AUC = 0.9853; CHASE_DB1: Acc = 0.9623, AUC = 0.9789) demonstrate the better performance of the proposed method compared to the state-of-the-art methods.

OntoSIDES: Ontology-based student progress monitoring on the national evaluation system of French Medical Schools. Publication date: Available online 19 March 2019 Source: Artificial Intelligence in Medicine Author(s): Olivier Palombi, Fabrice Jouanot, Nafissetou Nziengam, Behrooz Omidvar-Tehrani, Marie-Christine Rousset, Adam Sanchez Abstract We introduce OntoSIDES, the core of an ontology-based learning management system in Medicine, in which the educational content, the traces of students' activities and the correction of exams are linked and related to items of an official reference program in a unified RDF data model. OntoSIDES is an RDF knowledge base comprised of a lightweight domain ontology that serves as a pivot high-level vocabulary of the query interface with users, and of a dataset made of factual statements relating individual entities to classes and properties of the ontology. Thanks to an automatic mapping-based data materialization and rule-based data saturation, OntoSIDES contains around 8 millions triples to date, and provides an integrated access to useful information for student progress monitoring, using a powerful query language (namely SPARQL) allowing users to express their specific needs of data exploration and analysis. Since we do not expect end-users to master the raw syntax of SPARQL and to express directly complex queries in SPARQL, we have designed a set of parametrized queries that users can customize through a user-friendly interface.

Prediction of fetal state from the cardiotocogram recordings using neural network models Publication date: Available online 19 March 2019 Source: Artificial Intelligence in Medicine Author(s): Mohammad saber Iraji Abstract The combination of machine vision and soft computing approaches in the clinical decisions, using training data, can improve medical decisions and treatments. The cardiotocography (CTG) monitoring and uterine activity (UA) provides useful information about the condition of the fetus and the cesarean or natural delivery. The visual assessment by the pathologists takes a lot of time and may be incompatible. Therefore, creating a computer intelligent method to assess fetal wellbeing before the mother labour is very important. In this study, many diverse approaches are suggested for predicting fetal state classes based on artificial intelligence. The various topologies of multi-layer architecture of a sub-adaptive neuro fuzzy inference system (MLA-ANFIS) using multiple input features, neural networks (NN), deep stacked sparse auto-encoders (DSSAEs), and deep-ANFIS models are implemented on a CTG data set. Experimental results contributing to DSSAE are more accurate than other suggested techniques to predict fetal state. The proposed method achieved a sensitivity of 99.716, specificity of 97.500 and geometric mean of 98.602 with accuracy of 99.503.

Dynamic thresholding networks for schizophrenia diagnosis Publication date: Available online 18 March 2019 Source: Artificial Intelligence in Medicine Author(s): Hongliang Zou, Jian Yang Abstract Background and Objective Functional connectivity (FC) based on resting-state functional magnetic resonance imaging (rs-fMRI) is an effective approach to describe the neural interaction between distributed brain regions. Recent progress in neuroimaging study reported that the connection between regions is time-varying, which may enhance understanding of normal cognition and alterations that result from brain disorders. However, conventional sliding window based dynamic FC (DFC) analysis has several drawbacks, including arbitrary choice of window length, inaccurate descriptor of FC, and the fact that many spurious connections were included in the fully-connected networks due to noise. This study aims to develop an effective dynamic thresholding brain networks method to diagnose schizophrenia. Methods In this study, we proposed a time-varying window length DFC method based on dynamic time warping to construct brain functional networks. To further eliminate the influence of spurious connections caused by noise, orthogonal minimum spanning tree was applied in these networks to generate time-varying window length dynamic thresholding FC (TVWDTFC) networks. To validate the effectiveness of our proposed method, experiments were conducted on a dataset, which including 56 individuals with schizophrenia and 74 healthy controls. Results We achieved a classification accuracy of 0.8077 (p < 0.001, permutation test) using support vector machine. Experimental results demonstrated that the proposed method outperforms several state-of-the-art approaches, which verified the effectiveness of our proposed TVWDTFC method in schizophrenia diagnosis. Additionally, we also found that the selected discriminative features were mostly distributed in frontal, parietal, and limbic area. Conclusions The results suggest that our approach may be a promising tool for computer-aided diagnosis of schizophrenia.

Mining Heterogeneous Network for Drug Repositioning using Phenotypic Information Extracted from Social Media and Pharmaceutical Databases Publication date: Available online 9 March 2019 Source: Artificial Intelligence in Medicine Author(s): Christopher C. Yang, Mengnan Zhao Abstract Drug repositioning has drawn significant attention for drug development in pharmaceutical research and industry, because of its advantages in cost and time compared with the de novo drug development. The availability of biomedical databases and online health-related information, as well as the high-performance computing, empowers the development of computational drug repositioning methods. In this work, we developed a systematic approach that identifies repositioning drugs based on heterogeneous network mining using both pharmaceutical databases (PharmGKB and SIDER) and online health community (MedHelp). By utilizing adverse drug reactions (ADRs) as the intermediate, we constructed a heterogeneous health network containing drugs, diseases, and ADRs, and developed path-based heterogeneous network mining approaches for drug repositioning. Additionally, we investigated on how the data sources affect the performance on drug repositioning. Experiment results showed that combining both PharmKGB and MedHelp identified 479 repositioning drugs, which are more than the repositioning drugs discovered by other alternatives. In addition, 31% of the 479 of the discovered repositioning drugs were supported by evidence from PubMed.

Estimation of Echocardiogram parameters with the aid of Impedance Cardiography and Artificial Neural Networks Publication date: Available online 8 March 2019 Source: Artificial Intelligence in Medicine Author(s): Sudipta Ghosh, Bhabani Prasad Chattopadhyay, Ram Mohan Roy, Jayanta Mukherjee, Manjunatha Mahadevappa Abstract The advent of cardiovascular diseases as a disease of mass catastrophy, in recent years is alarming. It is expected to spread as an epidemic by 2030. Present methods of determining the health of one's heart include doppler based echocardiogram, MDCT (Multi Detector Computed Tomography), among various other invasive and non-invasive hemodynamic monitoring techniques. These methods require expert supervision and costly clinical set-ups, and cannot be employed by a common individual to perform a self diagnosis of one's cardiac health, unassisted. In this work, the authors propose a novel methodology using impedance cardiography (ICG), for the determination of a person's cardio-vascular health. The recorded ICG signal helps in extraction of features which are used for estimating parameters for cardiac health monitoring. The proposed methodology with the aid of artificial neural network is able to determine Stroke Volume (SV), Left Ventricular End Systolic Volume (LVESV), Left Ventricular End Diastolic Volume (LVEDV), Left Ventricular Ejection Fraction (LVEF), Iso Volumetric Contraction Time (IVCT), Iso Volumetric Relaxation Time (IVRT), Left Ventricular Ejection Time (LVET), Total Systolic Time (TST), Total Diastolic Time (TDT), and Myocardial Performance Index (MPI), with error margins of ±8.9%, ±3.8%, ±1.4%, ±7.8%, ±16.0%, ±9.0%, ±9.7%, ±6.9%, ±6.2%, and ±0.9%, respectively. The proposed methodology could be used in screening of precursors to cardiac ailments, and to keep a check on the cardio-vascular health.

Autonomous agents and multi-agent systems applied in healthcare Publication date: Available online 27 February 2019 Source: Artificial Intelligence in Medicine Author(s): Sara Montagna, Daniel Castro Silva, Pedro Henriques Abreu, Marcia Ito, Michael Ignaz Schumacher, Eloisa Vargiu

Real-time multi-agent systems for telerehabilitation scenarios Publication date: Available online 14 February 2019 Source: Artificial Intelligence in Medicine Author(s): Davide Calvaresi, Mauro Marinoni, Aldo Franco Dragoni, Roger Hilfiker, Michael Schumacher Abstract Telerehabilitation in older adults is most needed in the patient environments, rather than in formal ambulatories or hospitals. Supporting such practices brings significant advantages to patients, their family, formal and informal caregivers, clinicians, and researchers. This paper presents a focus group with experts in physiotherapy and telerehabilitation, debating on the requirements, current techniques and technologies developed to facilitate and enhance the effectiveness of telerehabilitation, and the still open challenges. Particular emphasis is given to (i) the body-parts requiring the most rehabilitation, (ii) the typical environments, initial causes, and general conditions, (iii) the values and parameters to be observed, (iv) common errors and limitations of current practices and technological solutions, and (v) the envisioned and desired technological support. Consequently, it has been performed a systematic review of the state of the art, investigating what types of systems and support currently cope with telerehabilitation practices and possible matches with the outcomes of the focus group. Technological solutions based on video analysis, wearable devices, robotic support, distributed sensing, and gamified telerehabilitation are examined. Particular emphasis is given to solutions implementing agent-based approaches, analyzing and discussing strength, limitations, and future challenges. By doing so, it has been possible to relate functional requirements expressed by professional physiotherapists and researchers, with the need for extending multi-agent systems (MAS) peculiarities at the sensing level in wearable solutions establishing new research challenges. In particular, to be employed in safety-critical cyber-physical scenarios with user-sensor and sensor-sensor interactions, MAS are requested to handle timing constraints, scarcity of resources and new communication means, crucial to providing real-time feedback and coaching. Therefore, MAS pillars such as the negotiation protocol and the agent's internal scheduler have been investigated, proposing solutions to achieve the aforementioned real-time compliance.