Search

This paper proposes the use of machine learning models to predict one's risk of having hypertension in the future using their routine health checkup data of their current and past visits to a health checkup center. The large-scale and high-dimensional dataset used in this study comes from MJ Health Research Foundation in Taiwan. The training data for models is separated into 5 folds and used to train 5 models in a 5-fold cross validation manner. While predicting the results for the test set, the voted result of 5 models is used as the final prediction. Experimental results show that our models achieve 69.59% of precision, 77.90% of recall, and 73.51% of F1-score, which outperforms a baseline using only the blood pressure of visitors' last visits. Experiments also show that a visitor who performs a health checkup more often can be predicted better, and models trained with selected important factors achieve better results than those trained with Framingham risk score. We also demonstrate the possibility of using our models to suggest visitors for weight control by adding virtual visits that assume their body weight can be reduced in the near future to model input. Experimental results show that around 5.48% of the people who are with high Body Mass Index of the true positive cases are rejudged as negative, and a rising trend appears when adding more virtual visits, which may be used to suggest visitors that controlling their body weight for a longer time lead to lower probability of having hypertension in the future., Competing Interests: The authors have declared that no competing interests exist., (Copyright: © 2024 Wang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.)

Background: Altered gut metabolites, especially short-chain fatty acids (SCFAs), in feces and plasma are observed in patients with Parkinson's disease (PD)., Objective: We aimed to investigate the colonic expression of two SCFA receptors, free fatty acid receptor (FFAR)2 and FFAR3, and gut barrier integrity in patients with PD and correlations with clinical severity., Methods: In this retrospective study, colonic biopsy specimens were collected from 37 PD patients and 34 unaffected controls. Of this cohort, 31 participants (14 PD, 17 controls) underwent a series of colon biopsies. Colonic expression of FFAR2, FFAR3, and the tight junction marker ZO-1 were assayed by immunofluorescence staining. The You Only Look Once (version 8, YOLOv8) algorithm was used for automated detection and segmentation of immunostaining signal. PD motor function was assessed with the Movement Disorder Society (MDS)-Unified Parkinson's Disease Rating Scale (UPDRS), and constipation was assessed using Rome-IV criteria., Results: Compared with controls, PD patients had significantly lower colonic expression of ZO-1 (p < 0.01) and FFAR2 (p = 0.01). On serial biopsy, colonic expression of FFAR2 and FFAR3 was reduced in the pre-motor stage before PD diagnosis (both p < 0.01). MDS-UPDRS motor scores did not correlate with colonic marker levels. Constipation severity negatively correlated with colonic ZO-1 levels (r = -0.49, p = 0.02)., Conclusions: Colonic expression of ZO-1 and FFAR2 is lower in PD patients compared with unaffected controls, and FFAR2 and FFAR3 levels decline in the pre-motor stage of PD. Our findings implicate a leaky gut phenomenon in PD and reinforce that gut metabolites may contribute to the process of PD., (© 2024. Fondazione Società Italiana di Neurologia.)

Background: Nigrosome-1 imaging has been used for assisting the diagnosis of Parkinson's disease (PD). We aimed to examine the diagnostic performance of loss of nigrosome-1 in PD and the correlation between the size of the nigrosome-1 and motor severity of PD., Methods: We included 237 patients with PD and 165 controls. The motor severity of PD was assessed with the Unified Parkinson's Disease Rating Scale (UPDRS) part III score and Hoehn-Yahr staging. The 3 or 1.5 Tesla susceptibility-weighted imaging combined with a deep-learning algorithm was applied for detecting the loss and the size of nigrosome-1. Clinical correlations and diagnostic performance of size of nigrosome-1 were also investigated., Results: The mean nigrosome-1 size was significantly smaller in PD patients than in controls (0.06 ± 0.07 cm 2 vs. 0.20 ± 0.05 cm 2 , P < 0.001). The area under the receiver operating characteristic curve (AUC) of the established model showed 0.94 accuracy (95% confidence interval [CI]: 0.87, 1.01, P < 0.01) in differentiating between the PD and control groups. Moreover, the partial loss of nigrosome-1 detected with SWI had an AUC of 0.96 in discriminating early-stage PD from controls (95% CI: 0.88, 1.02, P < 0.001). After adjusting for age, sex, disease duration, and levodopa equivalent daily dose, the estimated size of nigrosome-1 was negatively associated with the UPDRS part III motor score (ρ = -0.433, P < 0.001), but not with Mini-Mental State Examination scores (ρ = 0.006, P = 0.894)., Conclusions: The extent of loss and the size of nigrosome-1 may potentially assist in the diagnosis of PD. Nigrosome-1 size reflects the motor severity of PD., Competing Interests: Declaration of Competing Interest All authors declare no conflicts of interests., (Copyright © 2023 Elsevier B.V. All rights reserved.)

Alzheimer's disease (AD) progresses relentlessly from the preclinical to the dementia stage. The process begins decades before the diagnosis of dementia. Therefore, it is crucial to detect early manifestations to prevent cognitive decline. Recent advances in artificial intelligence help tackle the complex high-dimensional data encountered in clinical decision-making. In total, we recruited 206 subjects, including 69 cognitively unimpaired, 40 subjective cognitive decline (SCD), 34 mild cognitive impairment (MCI), and 63 AD dementia (ADD). We included 3 demographic, 1 clinical, 18 brain-image, and 3 plasma biomarker (Aß 1-42 , Aß 1-40 , and tau protein) features. We employed the linear discriminant analysis method for feature extraction to make data more distinguishable after dimension reduction. The sequential forward selection method was used for feature selection to identify the 12 best features for machine learning classifiers. We used both random forest and support vector machine as classifiers. The area under the receiver operative curve (AUROC) was close to 0.9 between diseased (combining ADD and MCI) and the controls. AUROC was higher than 0.85 between SCD and controls, 0.90 between MCI and SCD, and above 0.85 between ADD and MCI. We can differentiate between adjacent phases of the AD spectrum with blood biomarkers and brain MR images with the help of machine learning algorithms.

Background: Glaucoma is one of the major causes of blindness; it is estimated that over 110 million people will be affected by glaucoma worldwide by 2040. Research on glaucoma detection using deep learning technology has been increasing, but the diagnosis of glaucoma in a large population with high incidence of myopia remains a challenge. This study aimed to provide a decision support system for the automatic detection of glaucoma using fundus images, which can be applied for general screening, especially in areas of high incidence of myopia., Methods: A total of 1,155 fundus images were acquired from 667 individuals with a mean axial length of 25.60 ± 2.0 mm at the National Taiwan University Hospital, Hsinchu Br. These images were graded based on the findings of complete ophthalmology examinations, visual field test, and optical coherence tomography into three groups: normal (N, n = 596), pre-perimetric glaucoma (PPG, n = 66), and glaucoma (G, n = 493), and divided into a training-validation (N: 476, PPG: 55, G: 373) and test (N: 120, PPG: 11, G: 120) sets. A multimodal model with the Xception model as image feature extraction and machine learning algorithms [random forest (RF), support vector machine (SVM), dense neural network (DNN), and others] was applied., Results: The Xception model classified the N, PPG, and G groups with 93.9% of the micro-average area under the receiver operating characteristic curve (AUROC) with tenfold cross-validation. Although normal and glaucoma sensitivity can reach 93.51% and 86.13% respectively, the PPG sensitivity was only 30.27%. The AUROC increased to 96.4% in the N + PPG and G groups. The multimodal model with the N + PPG and G groups showed that the AUROCs of RF, SVM, and DNN were 99.56%, 99.59%, and 99.10%, respectively; The N and PPG + G groups had less than 1% difference. The test set showed an overall 3%-5% less AUROC than the validation results., Conclusion: The multimodal model had good AUROC while detecting glaucoma in a population with high incidence of myopia. The model shows the potential for general automatic screening and telemedicine, especially in Asia., Trial Registration: The study was approved by the Institutional Review Board of the National Taiwan University Hospital, Hsinchu Branch (no. NTUHHCB 108-025-E)., (© 2022. The Author(s).)

Hypomimia and voice changes are soft signs preceding classical motor disability in patients with Parkinson's disease (PD). We aim to investigate whether an analysis of acoustic and facial expressions with machine-learning algorithms assist early identification of patients with PD. We recruited 371 participants, including a training cohort (112 PD patients during "on" phase, 111 controls) and a validation cohort (74 PD patients during "off" phase, 74 controls). All participants underwent a smartphone-based, simultaneous recording of voice and facial expressions, while reading an article. Nine different machine learning classifiers were applied. We observed that integrated facial and voice features could discriminate early-stage PD patients from controls with an area under the receiver operating characteristic (AUROC) diagnostic value of 0.85. In the validation cohort, the optimal diagnostic value (0.90) maintained. We concluded that integrated biometric features of voice and facial expressions could assist the identification of early-stage PD patients from aged controls., (© 2022. The Author(s).)

Although in-hospital cardiac arrest is uncommon, it has a high mortality rate. Risk identification of at-risk patients is critical for post-cardiac arrest survival rates. Early warning scoring systems are generally used to identify hospitalized patients at risk of deterioration. However, these systems often require clinical data that are not always regularly measured. We developed a more accurate, machine learning-based model to predict clinical deterioration. The time series early warning score (TEWS) used only heart rate, systolic blood pressure, and respiratory data, which are regularly measured in general wards. We tested the performance of the TEWS in two tasks performed with data from the electronic medical records of 16,865 adult admissions and compared the results with those of other classifications. The TEWS detected more deteriorations with the same level of specificity as the different algorithms did when inputting vital signs data from 48 h before an event. Our framework improved in-hospital cardiac arrest prediction and demonstrated that previously obtained vital signs data can be used to identify at-risk patients in real-time. This model may be an alternative method for detecting patient deterioration., (© 2022. The Author(s).)

Background: Fatty liver disease (FLD) arises from the accumulation of fat in the liver and may cause liver inflammation, which, if not well controlled, may develop into liver fibrosis, cirrhosis, or even hepatocellular carcinoma., Objective: We describe the construction of machine-learning models for current-visit prediction (CVP), which can help physicians obtain more information for accurate diagnosis, and next-visit prediction (NVP), which can help physicians provide potential high-risk patients with advice to effectively prevent FLD., Methods: The large-scale and high-dimensional dataset used in this study comes from Taipei MJ Health Research Foundation in Taiwan. We used one-pass ranking and sequential forward selection (SFS) for feature selection in FLD prediction. For CVP, we explored multiple models, including k-nearest-neighbor classifier (KNNC), Adaboost, support vector machine (SVM), logistic regression (LR), random forest (RF), Gaussian naïve Bayes (GNB), decision trees C4.5 (C4.5), and classification and regression trees (CART). For NVP, we used long short-term memory (LSTM) and several of its variants as sequence classifiers that use various input sets for prediction. Model performance was evaluated based on two criteria: the accuracy of the test set and the intersection over union/coverage between the features selected by one-pass ranking/SFS and by domain experts. The accuracy, precision, recall, F-measure, and area under the receiver operating characteristic curve were calculated for both CVP and NVP for males and females, respectively., Results: After data cleaning, the dataset included 34,856 and 31,394 unique visits respectively for males and females for the period 2009-2016. The test accuracy of CVP using KNNC, Adaboost, SVM, LR, RF, GNB, C4.5, and CART was respectively 84.28%, 83.84%, 82.22%, 82.21%, 76.03%, 75.78%, and 75.53%. The test accuracy of NVP using LSTM, bidirectional LSTM (biLSTM), Stack-LSTM, Stack-biLSTM, and Attention-LSTM was respectively 76.54%, 76.66%, 77.23%, 76.84%, and 77.31% for fixed-interval features, and was 79.29%, 79.12%, 79.32%, 79.29%, and 78.36%, respectively, for variable-interval features., Conclusions: This study explored a large-scale FLD dataset with high dimensionality. We developed FLD prediction models for CVP and NVP. We also implemented efficient feature selection schemes for current- and next-visit prediction to compare the automatically selected features with expert-selected features. In particular, NVP emerged as more valuable from the viewpoint of preventive medicine. For NVP, we propose use of feature set 2 (with variable intervals), which is more compact and flexible. We have also tested several variants of LSTM in combination with two feature sets to identify the best match for male and female FLD prediction. More specifically, the best model for males was Stack-LSTM using feature set 2 (with 79.32% accuracy), whereas the best model for females was LSTM using feature set 1 (with 81.90% accuracy)., (©Cheng-Tse Wu, Ta-Wei Chu, Jyh-Shing Roger Jang. Originally published in JMIR Medical Informatics (https://medinform.jmir.org), 12.08.2021.)

The purpose of this study was to detect the presence of retinitis pigmentosa (RP) based on color fundus photographs using a deep learning model. A total of 1670 color fundus photographs from the Taiwan inherited retinal degeneration project and National Taiwan University Hospital were acquired and preprocessed. The fundus photographs were labeled RP or normal and divided into training and validation datasets (n = 1284) and a test dataset (n = 386). Three transfer learning models based on pre-trained Inception V3, Inception Resnet V2, and Xception deep learning architectures, respectively, were developed to classify the presence of RP on fundus images. The model sensitivity, specificity, and area under the receiver operating characteristic (AUROC) curve were compared. The results from the best transfer learning model were compared with the reading results of two general ophthalmologists, one retinal specialist, and one specialist in retina and inherited retinal degenerations. A total of 935 RP and 324 normal images were used to train the models. The test dataset consisted of 193 RP and 193 normal images. Among the three transfer learning models evaluated, the Xception model had the best performance, achieving an AUROC of 96.74%. Gradient-weighted class activation mapping indicated that the contrast between the periphery and the macula on fundus photographs was an important feature in detecting RP. False-positive results were mostly obtained in cases of high myopia with highly tessellated retina, and false-negative results were mostly obtained in cases of unclear media, such as cataract, that led to a decrease in the contrast between the peripheral retina and the macula. Our model demonstrated the highest accuracy of 96.00%, which was comparable with the average results of 81.50%, of the other four ophthalmologists. Moreover, the accuracy was obtained at the same level of sensitivity (95.71%), as compared to an inherited retinal disease specialist. RP is an important disease, but its early and precise diagnosis is challenging. We developed and evaluated a transfer-learning-based model to detect RP from color fundus photographs. The results of this study validate the utility of deep learning in automating the identification of RP from fundus photographs., (© 2021. Society for Imaging Informatics in Medicine.)

Easily accessible biomarkers for Alzheimer's disease (AD), Parkinson's disease (PD), frontotemporal dementia (FTD), and related neurodegenerative disorders are urgently needed in an aging society to assist early-stage diagnoses. In this study, we aimed to develop machine learning algorithms using the multiplex blood-based biomarkers to identify patients with different neurodegenerative diseases. Plasma samples ( n = 377) were obtained from healthy controls, patients with AD spectrum (including mild cognitive impairment (MCI)), PD spectrum with variable cognitive severity (including PD with dementia (PDD)), and FTD. We measured plasma levels of amyloid-beta 42 (Aβ42), Aβ40, total Tau, p-Tau181, and α-synuclein using an immunomagnetic reduction-based immunoassay. We observed increased levels of all biomarkers except Aβ40 in the AD group when compared to the MCI and controls. The plasma α-synuclein levels increased in PDD when compared to PD with normal cognition. We applied machine learning-based frameworks, including a linear discriminant analysis (LDA), for feature extraction and several classifiers, using features from these blood-based biomarkers to classify these neurodegenerative disorders. We found that the random forest (RF) was the best classifier to separate different dementia syndromes. Using RF, the established LDA model had an average accuracy of 76% when classifying AD, PD spectrum, and FTD. Moreover, we found 83% and 63% accuracies when differentiating the individual disease severity of subgroups in the AD and PD spectrum, respectively. The developed LDA model with the RF classifier can assist clinicians in distinguishing variable neurodegenerative disorders.

Vector-valued neural learning has emerged as a promising direction in deep learning recently. Traditionally, training data for neural networks (NNs) are formulated as a vector of scalars; however, its performance may not be optimal since associations among adjacent scalars are not modeled. In this article, we propose a new vector neural architecture called the Arbitrary BIlinear Product NN (ABIPNN), which processes information as vectors in each neuron, and the feedforward projections are defined using arbitrary bilinear products. Such bilinear products can include circular convolution, 7-D vector product, skew circular convolution, reversed-time circular convolution, or other new products that are not seen in the previous work. As a proof-of-concept, we apply our proposed network to multispectral image denoising and singing voice separation. Experimental results show that ABIPNN obtains substantial improvements when compared to conventional NNs, suggesting that associations are learned during training.

A novel Bisphenol A (4,4'-isopropylidenediphenol, BPA) sensor was developed harnessing an electrochemical platform comprising a layer-by-layer assembled reduced graphene oxide (rGO) electrode and a designer probe specifically recognizing BPA. The BPA detection probe, a recombinant protein (LacI-BPA), was constructed by fusing a disulfide-constrained high affinity BPA binding peptide (CKSLENSYC) to the C-terminus of Lac repressor (LacI). Following expression and purification, the LacI-BPA was heat-denatured on-purpose to facilitate its direct adhesion on the rGO electrode surface via pi-stacking interaction. When the performance of the fabricated BPA sensor (LacI-BPA/rGO) was assessed by electrochemical impedance spectroscopy (EIS), it showed a wide linear dynamic range of BPA detection spanning from 100 fM to 10nM. Moreover, our BPA sensor exhibited negligible cross reactivity to BPA analogs such as Bisphenol S (BPS) and Bisphenol F (BPF) and almost complete spike recovery of BPA from plastic extracts containing various potential interferents. With these merits, the BPA sensor developed in the present study is expected to find practical application in selective and sensitive detection of BPA from diverse sample solutions., (Copyright © 2015 Elsevier B.V. All rights reserved.)

Transition metal oxide (TMO) thin films have been exploited as interlayers for charge extraction between electrodes and active layers in organic photovoltaic (OPV) devices. Additionally, graphene-electrode-based OPVs have received considerable attention as a means to enhance device stability. However, the film deposition process of a TMO thin-film layer onto the graphene electrode is highly restricted owing to the hydrophobic nature of the graphene surface; thus, the preparation of the device should rely on a vacuum process that is incompatible with solution processing. In this study, we present a novel means for creating a thin tungsten oxide (WO3 ) interlayer on a graphene electrode by employing an engineered biotemplate of M13 viruses, whereby nondestructive functionalization of the graphene and uniform synthesis of a WO3 thin interlayer are concurrently achieved. As a result, the incorporated virus-templated WO3 interlayer exhibited solar-conversion efficiency that was 20 % higher than that of conventional OPVs based on the use of a (3,4-ethylenedioxythiophene):poly(styrenesulfonate) (, Pedot: PSS) interlayer. Notably, bilayer-structured OPVs with synergistically integrated WO3 /PEDOT:PSS achieved >60 % enhancement in device performance., (© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.)

In this work, we demonstrated tunable p- and/or n-type doping of chemical vapor deposition-grown graphene with the use of protein bovine serum albumin (BSA) as a dopant. BSA undergoes protonation or deprotonation reaction subject to solution pH, thereby acting as either an electron donor or an electron acceptor on the graphene surface layered with denatured BSA through π-stacking interaction. This direct annealing of graphene with denatured BSA of amphoteric nature rendered facilitated fabrication of a p- and/or n-type graphene transistor by modulating pH-dependent net charges of the single dopant. Following AFM confirmation of the BSA/graphene interface assembly, the carrier transport properties of BSA-doped graphene transistors were assessed by I-V measurement and Raman spectra to show effective charge modulation of the graphene enabled by BSA doping at various pH conditions. The protein-mediated bipolar doping of graphene demonstrated in our work is simple, scalable, and straightforward; the proposed scheme is therefore expected to provide a useful alternative for fabricating graphene transistors of novel properties and promote their implementation in practice.

Titanium-tungsten oxide composites with greatly enhanced photocatalytic activity were synthesized by lysozyme-mediated biomineralization. It was shown for the first time that simple control of the onset of biomineralization could enable fine tuning of the composition and crystallinity of the composites to determine their photocatalytic performance.

Graphene-based electrochemical impedance sensors have recently received much attention due to their outstanding sensing capability and economic viability. In this study, we present a novel means of constructing an impedance sensing platform via harnessing intrinsic π-stacking interactions between probe protein molecules and reduced graphene oxide (RGO) substrate, obviating the need for introducing external chemical groups often required for covalent anchoring of the probes. To achieve this goal, protein molecules used as a probe were denatured to render their hydrophobic residues exposed in order to facilitate their direct π-stacking interactions with the surface of RGO nanosheets. The protein molecules in denatured form, which would otherwise have difficulty in undergoing π-stacking interactions with the RGO surface, were found to uniformly cover the RGO nanosheets at high density, conducive to providing a graphene-based impedance sensing platform capable of detecting a probe-specific analyte at high sensitivity. The proof-of-concept performance of thus-constructed RGO-based impedance sensors was demonstrated via selective detection of biological binding events of antigen-antibody reaction at a femtomolar range. Notably, since the π-stacking interaction can occur on the entire RGO surface, it can desirably exclude a backfill process indispensable for the conventional biosensors to suppress background noise signals. Since the procedure of π-stacking mediated direct deposition of on-purpose denatured protein probes onto the RGO surface is facile and straightforward, the proposed strategy is anticipated to extend its applicability for fabrication of high performance graphene-based bio or chemical sensors.

A highly specific lead-binding peptide ThrAsnThrLeuSerAsnAsn was displayed on Escherichia coli, and lead adsorption characteristics of the recombinant bacteria were investigated. Cell surface-displayed peptide was expressed under the control of an arabinose promoter using outer membrane protein C (OmpC(t)) as an anchoring motif. The optimal induction period and arabinose concentration for the expression of peptide-fused OmpC(t) were determined to be 2 h and 0.001 g/L, respectively. Selective adsorption of Pb(2+) onto recombinant cells was verified with individual or combinatory use of four metal ions, Pb(2+), Ni(2+), Co(2+), and Cu(2+); the amount of bound Pb(2+) onto the biosorbents was significantly higher than the other metal ions. The adsorption isotherm of recombinant cells for Pb(2+) followed the Langmuir isotherm with a maximum adsorption loading (q (max)) of 526 μmol/g dry cell weight.

Using cyclic constrained TiO(2) binding peptides STB1 (CHKKPSKSC), RSTB1 (CHRRPSRSC) and linear peptide LSTB1 (AHKKPSKSA), it was shown that while affinity of the peptide to TiO(2) is essential to enable TiO(2) biomineralization, other factors such as biomineralization kinetics and peptide local structure need to be considered to predict biomineralization efficacy. Cyclic and linear TiO(2) binding peptides show significantly different biomineralization activities. Cyclic STB1 and RSTB1 could induce TiO(2) precipitation in the presence of titanium(IV)-bis-ammonium-lactato-dihydroxide (TiBALDH) precursor in water or tris buffer at pH 8. In contrast, linear LSTB1 was unable to mineralize TiO(2) under the same experimental conditions despite its high affinity to TiO(2) comparable with STB1 and/or RSTB1. LSTB1 being a flexible molecule could not render the stable condensation of TiBALDH precursor to form TiO(2) particles. However, in the presence of phosphate buffer ions, the structure of LSTB1 is stabilized, leading to efficient condensation of TiBALDH and TiO(2) particle formation. This study demonstrates that peptide-mediated TiO(2) mineralization is governed by a complicated interplay of peptide sequence, local structure, kinetics and the presence of mineralizing aider such as phosphate ions., (Copyright © 2012 Elsevier Inc. All rights reserved.)

It is shown that, under some minor restrictions, the functional behavior of radial basis function networks (RBFNs) and that of fuzzy inference systems are actually equivalent. This functional equivalence makes it possible to apply what has been discovered (learning rule, representational power, etc.) for one of the models to the other, and vice versa. It is of interest to observe that two models stemming from different origins turn out to be functionally equivalent.

A generalized control strategy that enhances fuzzy controllers with self-learning capability for achieving prescribed control objectives in a near-optimal manner is presented. This methodology, termed temporal backpropagation, is model-sensitive in the sense that it can deal with plants that can be represented in a piecewise-differentiable format, such as difference equations, neural networks, GMDH structures, and fuzzy models. Regardless of the numbers of inputs and outputs of the plants under consideration, the proposed approach can either refine the fuzzy if-then rules of human experts or automatically derive the fuzzy if-then rules if human experts are not available. The inverted pendulum system is employed as a testbed to demonstrate the effectiveness of the proposed control scheme and the robustness of the acquired fuzzy controller.