Evolution of the Stethoscope: Advances with the Adoption of Machine Learning and Development of Wearable Devices.

The stethoscope has long been used for the examination of patients, but the importance of auscultation has declined due to its several limitations and the development of other diagnostic tools. However, auscultation is still recognized as a primary diagnostic device because it is non-invasive and provides valuable information in real-time. To supplement the limitations of existing stethoscopes, digital stethoscopes with machine learning (ML) algorithms have been developed. Thus, now we can record and share respiratory sounds and artificial intelligence (AI)-assisted auscultation using ML algorithms distinguishes the type of sounds. Recently, the demands for remote care and non-face-to-face treatment diseases requiring isolation such as coronavirus disease 2019 (COVID-19) infection increased. To address these problems, wireless and wearable stethoscopes are being developed with the advances in battery technology and integrated sensors. This review provides the history of the stethoscope and classification of respiratory sounds, describes ML algorithms, and introduces new auscultation methods based on AI-assisted analysis and wireless or wearable stethoscopes.

Prediction of disorders with significant coronary lesions using machine learning in patients admitted with chest symptom.

BACKGROUND: The early prediction of significant coronary artery lesion, including coronary vasospasm, have yet to be studied. It is essential to discern the disorders with significant coronary lesions (SCDs) requiring coronary angiography from mimicking disease. We aimed to determine which of all clinical variables were more important using conventional logistic regression (cLR) and machine learning (ML). MATERIALS: Of 3382 patients with chest pain/discomfort or dyspnea in whom CAG was performed, 1893 were included. All clinical data were divided as follows (i): Demographics, history, and physical examination; (ii): (i) plus electrocardiography; and (iii): (ii) plus echocardiography, and analyzed by cLR and ML. RESULTS: In multivariable analysis via cLR, the AUC and accuracy of the model using the final 20 variables were 0.795 and 72.62%, respectively. In multivariable analysis via ML, the best AUCs in the internal validation were 0.8 with (i), 0.81 with (ii), 0.83 with (iii), and in external validation, the best AUCs were 0.71 with (i), 0.74 with (ii), and 0.79 with (iii). The best AUCs and accuracy of the fittest model including 21 importance variables by ML were 0.81 and 72.48% in internal validation; and 0.75 and 70.5% in external validation, respectively. The importance variables in ML and cLR were similar, but slightly different and the additional discriminators via ML were found. CONCLUSION: The assessment using the fittest importance variables can assist physicians in differentiating mimicking diseases in which coronary angiography may not be required in patients suspected of having acute coronary syndrome in emergency department.

The coming era of a new auscultation system for analyzing respiratory sounds.

Auscultation with stethoscope has been an essential tool for diagnosing the patients with respiratory disease. Although auscultation is non-invasive, rapid, and inexpensive, it has intrinsic limitations such as inter-listener variability and subjectivity, and the examination must be performed face-to-face. Conventional stethoscope could not record the respiratory sounds, so it was impossible to share the sounds. Recent innovative digital stethoscopes have overcome the limitations and enabled clinicians to store and share the sounds for education and discussion. In particular, the recordable stethoscope made it possible to analyze breathing sounds using artificial intelligence, especially based on neural network. Deep learning-based analysis with an automatic feature extractor and convoluted neural network classifier has been applied for the accurate analysis of respiratory sounds. In addition, the current advances in battery technology, embedded processors with low power consumption, and integrated sensors make possible the development of wearable and wireless stethoscopes, which can help to examine patients living in areas of a shortage of doctors or those who need isolation. There are still challenges to overcome, such as the analysis of complex and mixed respiratory sounds and noise filtering, but continuous research and technological development will facilitate the transition to a new era of a wearable and smart stethoscope.

Machine learning-based survival rate prediction of Korean hepatocellular carcinoma patients using multi-center data.

AIM: To predict survival time of Korean hepatocellular carcinoma (HCC) patients using multi-center data as a foundation for the development of a predictive artificial intelligence model according to treatment methods based on machine learning. METHODS: Data of patients who underwent treatment for HCC from 2008 to 2015 was provided by Korean Liver Cancer Study Group and Korea Central Cancer Registry. A total of 10,742 patients with HCC were divided into two groups, with Group I (2920 patients) confirmed on biopsy and Group II (5562 patients) diagnosed as HCC according to HCC diagnostic criteria as outlined in Korean Liver Cancer Association guidelines. The data were modeled according to features of patient clinical characteristics. Features effective in predicting survival rate were analyzed retrospectively. Various machine learning methods were used. RESULTS: Target was overall survival time, which divided into approximately 60 months (= /< 60 m, > 60 m). Target distribution in Group I (total 514 samples) was 28.8%: (148 samples) less than 60 months, 71.2% (366 samples) greater than 60 months, and in Group II (total 757 samples) was 66.6% (504 samples) less than 60 months, 33.4% (253 samples) greater than 60 months. Using NG Boost method, its accuracy was 83%, precision 84%, sensitivity 95%, and F1 score 89% for more than 60 months survival time in Group I with surgical resection. Moreover, its accuracy was 79%, precision 82%, sensitivity 87%, and F1 score 84% for less than 60 months survival time in Group II with TACE. The feature importance with gain criterion indicated that pathology, portal vein invasion, surgery, metastasis, and needle biopsy features could be explained as important factors for prediction in case of biopsy (Group I). CONCLUSION: By developing a predictive model using machine learning algorithms to predict prognosis of HCC patients, it is possible to project optimized treatment by case according to liver function and tumor status.

Body Fat Plays an Important Role in of Bioimpedance Spectroscopy-Based Dry Weight Measurement Error for Patients with Hemodialysis.

Accurate dry weight (DW) estimation is important for hemodialysis patients. Although bioimpedance spectroscopy (BIS) is commonly used to measure DW, the BIS-based DW frequently differs from the clinical DW. We analyzed the characteristics of patients whose BIS-based DWs were over- and underestimated. In this retrospective cohort study, we evaluated 1555 patients undergoing maintenance hemodialysis in Chungnam National University Hospital. The gap (DWCP-BIS) was calculated by comparing the BIS and clinical DWs. We analyzed the clinical characteristics of patients with positive (n = 835) and negative (n = 720) gaps. Compared with other patients, the DWCP-BIS-positive group had higher extracellular water (ECW) level and extracellular/intracellular water index (E/I) and had lower weight, body mass index (BMI), lean tissue index (LTI), fat tissue index (FTI), fat mass (FAT), and adipose tissue mass (ATM). The DWCP-BIS-negative group exhibited elevated BMI, FTI, FAT, and ATM; however, it had lower height, ECW, and E/I. Linear regression analysis revealed that FAT significantly predicted DWCP accuracy. The clinical DW of patients with a low fat mass tended to be underestimated, while the clinical DW of patients with comparatively large fat reserves tended to be overestimated. These characteristics will aid in the reduction of BIS-based DW errors.

Respiratory sound classification for crackles, wheezes, and rhonchi in the clinical field using deep learning.

Auscultation has been essential part of the physical examination; this is non-invasive, real-time, and very informative. Detection of abnormal respiratory sounds with a stethoscope is important in diagnosing respiratory diseases and providing first aid. However, accurate interpretation of respiratory sounds requires clinician's considerable expertise, so trainees such as interns and residents sometimes misidentify respiratory sounds. To overcome such limitations, we tried to develop an automated classification of breath sounds. We utilized deep learning convolutional neural network (CNN) to categorize 1918 respiratory sounds (normal, crackles, wheezes, rhonchi) recorded in the clinical setting. We developed the predictive model for respiratory sound classification combining pretrained image feature extractor of series, respiratory sound, and CNN classifier. It detected abnormal sounds with an accuracy of 86.5% and the area under the ROC curve (AUC) of 0.93. It further classified abnormal lung sounds into crackles, wheezes, or rhonchi with an overall accuracy of 85.7% and a mean AUC of 0.92. On the other hand, as a result of respiratory sound classification by different groups showed varying degree in terms of accuracy; the overall accuracies were 60.3% for medical students, 53.4% for interns, 68.8% for residents, and 80.1% for fellows. Our deep learning-based classification would be able to complement the inaccuracies of clinicians' auscultation, and it may aid in the rapid diagnosis and appropriate treatment of respiratory diseases.

Swimming fluctuations of micro-organisms due to heterogeneous microstructure.

Swimming microorganisms in biological complex fluids may be greatly influenced by heterogeneous media and microstructure with length scales comparable to the organisms. A fundamental effect of swimming in a heterogeneous rather than homogeneous medium is that variations in local environments lead to swimming velocity fluctuations. Here we examine long-range hydrodynamic contributions to these fluctuations using a Najafi-Golestanian swimmer near spherical and filamentous obstacles. We find that forces on microstructures determine changes in swimming speed. For macroscopically isotropic networks, we also show how the variance of the fluctuations in swimming speeds are related to density and orientational correlations in the medium.

Energy variational analysis of ions in water and channels: Field theory for primitive models of complex ionic fluids.

Ionic solutions are mixtures of interacting anions and cations. They hardly resemble dilute gases of uncharged noninteracting point particles described in elementary textbooks. Biological and electrochemical solutions have many components that interact strongly as they flow in concentrated environments near electrodes, ion channels, or active sites of enzymes. Interactions in concentrated environments help determine the characteristic properties of electrodes, enzymes, and ion channels. Flows are driven by a combination of electrical and chemical potentials that depend on the charges, concentrations, and sizes of all ions, not just the same type of ion. We use a variational method EnVarA (energy variational analysis) that combines Hamilton's least action and Rayleigh's dissipation principles to create a variational field theory that includes flow, friction, and complex structure with physical boundary conditions. EnVarA optimizes both the action integral functional of classical mechanics and the dissipation functional. These functionals can include entropy and dissipation as well as potential energy. The stationary point of the action is determined with respect to the trajectory of particles. The stationary point of the dissipation is determined with respect to rate functions (such as velocity). Both variations are written in one Eulerian (laboratory) framework. In variational analysis, an "extra layer" of mathematics is used to derive partial differential equations. Energies and dissipations of different components are combined in EnVarA and Euler-Lagrange equations are then derived. These partial differential equations are the unique consequence of the contributions of individual components. The form and parameters of the partial differential equations are determined by algebra without additional physical content or assumptions. The partial differential equations of mixtures automatically combine physical properties of individual (unmixed) components. If a new component is added to the energy or dissipation, the Euler-Lagrange equations change form and interaction terms appear without additional adjustable parameters. EnVarA has previously been used to compute properties of liquid crystals, polymer fluids, and electrorheological fluids containing solid balls and charged oil droplets that fission and fuse. Here we apply EnVarA to the primitive model of electrolytes in which ions are spheres in a frictional dielectric. The resulting Euler-Lagrange equations include electrostatics and diffusion and friction. They are a time dependent generalization of the Poisson-Nernst-Planck equations of semiconductors, electrochemistry, and molecular biophysics. They include the finite diameter of ions. The EnVarA treatment is applied to ions next to a charged wall, where layering is observed. Applied to an ion channel, EnVarA calculates a quick transient pile-up of electric charge, transient and steady flow through the channel, stationary "binding" in the channel, and the eventual accumulation of salts in "unstirred layers" near channels. EnVarA treats electrolytes in a unified way as complex rather than simple fluids. Ad hoc descriptions of interactions and flow have been used in many areas of science to deal with the nonideal properties of electrolytes. It seems likely that the variational treatment can simplify, unify, and perhaps derive and improve those descriptions.