Use of artificial intelligence and I-Score for prediction of recurrence before catheter ablation of atrial fibrillation

BACKGROUND

Based solely on pre-ablation characteristics, previous risk scores have demonstrated variable predictive performance. This study aimed to predict the recurrence of AF after catheter ablation by using artificial intelligence (AI)-enabled pre-ablation computed tomography (PVCT) images and pre-ablation clinical data.

METHODS

A total of 638 drug-refractory paroxysmal atrial fibrillation (AF) patients undergone ablation were recruited. For model training, we used left atria (LA) acquired from pre-ablation PVCT slices (126,288 images). A total of 29 clinical variables were collected before ablation, including baseline characteristics, medical histories, laboratory results, transthoracic echocardiographic parameters, and 3D reconstructed LA volumes. The I-Score was applied to select variables for model training. For the prediction of one-year AF recurrence, PVCT deep-learning and clinical variable machine-learning models were developed. We then applied machine learning to ensemble the PVCT and clinical variable models.

RESULTS

The PVCT model achieved an AUC of 0.63 in the test set. Various combinations of clinical variables selected by I-Score can yield an AUC of 0.72, which is significantly better than all variables or features selected by nonparametric statistics (AUCs of 0.66 to 0.69). The ensemble model (PVCT images and clinical variables) significantly improved predictive performance up to an AUC of 0.76 (sensitivity of 86.7% and specificity of 51.0%).

CONCLUSIONS

Before ablation, AI-enabled PVCT combined with I-Score features was applicable in predicting recurrence in paroxysmal AF patients. Based on all possible predictors, the I-Score is capable of identifying the most influential combination.

Precision Identification of Locally Advanced Rectal Cancer in Denoised CT Scans Using EfficientNet and Voting System Algorithms

BACKGROUND AND OBJECTIVE

Local advanced rectal cancer (LARC) poses significant treatment challenges due to its location and high recurrence rates. Accurate early detection is vital for treatment planning. With magnetic resonance imaging (MRI) being resource-intensive, this study explores using artificial intelligence (AI) to interpret computed tomography (CT) scans as an alternative, providing a quicker, more accessible diagnostic tool for LARC.

METHODS

In this retrospective study, CT images of 1070 T3-4 rectal cancer patients from 2010 to 2022 were analyzed. AI models, trained on 739 cases, were validated using two test sets of 134 and 197 cases. By utilizing techniques such as nonlocal mean filtering, dynamic histogram equalization, and the EfficientNetB0 algorithm, we identified images featuring characteristics of a positive circumferential resection margin (CRM) for the diagnosis of locally advanced rectal cancer (LARC). Importantly, this study employs an innovative approach by using both hard and soft voting systems in the second stage to ascertain the LARC status of cases, thus emphasizing the novelty of the soft voting system for improved case identification accuracy. The local recurrence rates and overall survival of the cases predicted by our model were assessed to underscore its clinical value.

RESULTS

The AI model exhibited high accuracy in identifying CRM-positive images, achieving an area under the curve (AUC) of 0.89 in the first test set and 0.86 in the second. In a patient-based analysis, the model reached AUCs of 0.84 and 0.79 using a hard voting system. Employing a soft voting system, the model attained AUCs of 0.93 and 0.88, respectively. Notably, AI-identified LARC cases exhibited a significantly higher five-year local recurrence rate and displayed a trend towards increased mortality across various thresholds. Furthermore, the model's capability to predict adverse clinical outcomes was superior to those of traditional assessments.

CONCLUSION

AI can precisely identify CRM-positive LARC cases from CT images, signaling an increased local recurrence and mortality rate. Our study presents a swifter and more reliable method for detecting LARC compared to traditional CT or MRI techniques.

Axial length, more than aging, decreases the thickness and superficial vessel density of retinal nerve fiber layer in non-glaucomatous eyes

PURPOSE

This study seeks to build a normative database for the vessel density of the superficial retina (SVD) and evaluate how changes and trends in the retinal microvasculature may be influenced by age and axial length (AL) in non-glaucomatous eyes, as measured with optical coherence tomography angiography (OCTA).

METHODS

We included 500 eyes of 290 healthy subjects visiting a county hospital. Each participant underwent comprehensive ophthalmological examinations and OCTA to measure the SVD and thickness of the macular and peripapillary areas. To analyze correlations between SVD and age or AL, multivariable linear regression models with generalized estimating equations were applied.

RESULTS

Age was negatively correlated with the SVD of the superior, central, and inferior macular areas and the superior peripapillary area, with a decrease rate of 1.06%, 1.36%, 0.84%, and 0.66% per decade, respectively. However, inferior peripapillary SVD showed no significant correlation with age. AL was negatively correlated with the SVD of the inferior macular area and the superior and inferior peripapillary areas, with coefficients of -0.522%/mm, -0.733%/mm, and -0.664%/mm, respectively. AL was also negatively correlated with the thickness of the retinal nerve fiber layer and inferior ganglion cell complex (p = 0.004).

CONCLUSION

Age and AL were the two main factors affecting changes in SVD. Furthermore, AL, a relative term to represent the degree of myopia, had a greater effect than age and showed a more significant effect on thickness than on SVD. This relationship has important implications because myopia is a significant issue in modern cities.

Improving deep-learning electrocardiogram classification with an effective coloring method

Cardiovascular diseases, particularly arrhythmias, remain a leading cause of mortality worldwide. Electrocardiogram (ECG) analysis plays a pivotal role in cardiovascular disease diagnosis. Although previous studies have focused on waveform analysis and model training, integrating additional clinical information, especially demographic data, remains challenging. In this study, we present an innovative approach to ECG classification by incorporating demographic information from patients' medical histories through a colorization technique. Our proposed method maps demographic features onto the (R, G, B) color space through normalized scaling. Each demographic feature corresponds to a distinct color, allowing for different ECG leads to be colored. This approach preserves the relationships between data by maintaining the color correlations in the statistical features, enhancing ECG analytics and supporting precision medicine. We conducted experiments with PTB-XL dataset and achieved 1%-6% improvements in the area under the receiving operator characteristic curve performance compared with other methods for various classification problems. Notably, our method excelled in multiclass and challenging classification tasks. The combined use of color features and the original waveform shape features enhanced prediction accuracy for various deep learning models. Our findings suggest that colorization is a promising avenue for advancing ECG classification and diagnosis, contributing to improved prediction and diagnosis of cardiovascular diseases and ultimately enhancing clinical outcomes.

Specific patterns and potential risk factors to predict 3-year risk of death among non-cancer patients with advanced chronic kidney disease by machine learning

Chronic kidney disease (CKD) is a major public health concern. But there are limited machine learning studies on non-cancer patients with advanced CKD, and the results of machine learning studies on cancer patients with CKD may not apply directly on non-cancer patients. We aimed to conduct a comprehensive investigation of risk factors for a 3-year risk of death among non-cancer advanced CKD patients with an estimated glomerular filtration rate < 60.0 mL/min/1.73m2 by several machine learning algorithms. In this retrospective cohort study, we collected data from in-hospital and emergency care patients from 2 hospitals in Taiwan from 2009 to 2019, including their international classification of disease at admission and laboratory data from the hospital's electronic medical records (EMRs). Several machine learning algorithms were used to analyze the potential impact and degree of influence of each factor on mortality and survival. Data from 2 hospitals in northern Taiwan were collected with 6565 enrolled patients. After data cleaning, 26 risk factors and approximately 3887 advanced CKD patients from Shuang Ho Hospital were used as the training set. The validation set contained 2299 patients from Taipei Medical University Hospital. Predictive variables, such as albumin, PT-INR, and age, were the top 3 significant risk factors with paramount influence on mortality prediction. In the receiver operating characteristic curve, the random forest had the highest values for accuracy above 0.80. MLP, and Adaboost had better performance on sensitivity and F1-score compared to other methods. Additionally, SVM with linear kernel function had the highest specificity of 0.9983, while its sensitivity and F1-score were poor. Logistic regression had the best performance, with an area under the curve of 0.8527. Evaluating Taiwanese advanced CKD patients' EMRs could provide physicians with a good approximation of the patients' 3-year risk of death by machine learning algorithms.

Developing a Machine Learning Algorithm to Predict the Probability of Medical Staff Work Mode Using Human-Smartphone Interaction Patterns: Algorithm Development and Validation Study

BACKGROUND

Traditional methods for investigating work hours rely on an employee's physical presence at the worksite. However, accurately identifying break times at the worksite and distinguishing remote work outside the worksite poses challenges in work hour estimations. Machine learning has the potential to differentiate between human-smartphone interactions at work and off work.

OBJECTIVE

In this study, we aimed to develop a novel approach called "probability in work mode," which leverages human-smartphone interaction patterns and corresponding GPS location data to estimate work hours.

METHODS

To capture human-smartphone interactions and GPS locations, we used the "Staff Hours" app, developed by our team, to passively and continuously record participants' screen events, including timestamps of notifications, screen on or off occurrences, and app usage patterns. Extreme gradient boosted trees were used to transform these interaction patterns into a probability, while 1-dimensional convolutional neural networks generated successive probabilities based on previous sequence probabilities. The resulting probability in work mode allowed us to discern periods of office work, off-work, breaks at the worksite, and remote work.

RESULTS

Our study included 121 participants, contributing to a total of 5503 person-days (person-days represent the cumulative number of days across all participants on which data were collected and analyzed). The developed machine learning model exhibited an average prediction performance, measured by the area under the receiver operating characteristic curve, of 0.915 (SD 0.064). Work hours estimated using the probability in work mode (higher than 0.5) were significantly longer (mean 11.2, SD 2.8 hours per day) than the GPS-defined counterparts (mean 10.2, SD 2.3 hours per day; P<.001). This discrepancy was attributed to the higher remote work time of 111.6 (SD 106.4) minutes compared to the break time of 54.7 (SD 74.5) minutes.

CONCLUSIONS

Our novel approach, the probability in work mode, harnessed human-smartphone interaction patterns and machine learning models to enhance the precision and accuracy of work hour investigation. By integrating human-smartphone interactions and GPS data, our method provides valuable insights into work patterns, including remote work and breaks, offering potential applications in optimizing work productivity and well-being.

Dynamically Synthetic Images for Federated Learning of medical images

BACKGROUND

To develop deep learning models for medical diagnosis, it is important to collect more medical data from several medical institutions. Due to the regulations for privacy concerns, it is infeasible to collect data from various medical institutions to one institution for centralized learning. Federated Learning (FL) provides a feasible approach to jointly train the deep learning model with data stored in various medical institutions instead of collected together. However, the resulting FL models could be biased towards institutions with larger training datasets.

METHODOLOGY

In this study, we propose the applicable method of Dynamically Synthetic Images for Federated Learning (DSIFL) that aims to integrate the information of local institutions with heterogeneous types of data. The main technique of DSIFL is to develop a synthetic method that can dynamically adjust the number of synthetic images similar to local data that are misclassified by the current model. The resulting global model can handle the diversity in heterogeneous types of data collected in local medical institutions by including the training of synthetic images similar to misclassified cases in local collections.

RESULTS

In model performance evaluation metrics, we focus on the accuracy of each client's dataset. Finally, the accuracy of the model of DSIFL in the experiments can achieve the higher accuracy of the FL approach.

CONCLUSION

In this study, we propose the framework of DSIFL that achieves improvements over the conventional FL approach. We conduct empirical studies with two kinds of medical images. We compare the performance by variants of FL vs. DSIFL approaches. The performance by individual training is used as the baseline, whereas the performance by centralized learning is used as the target for the comparison studies. The empirical findings suggest that the DSIFL has improved performance over the FL via the technique of dynamically synthetic images in training.

Feature-aware unsupervised lesion segmentation for brain tumor images using fast data density functional transform

We demonstrate that isomorphically mapping gray-level medical image matrices onto energy spaces underlying the framework of fast data density functional transform (fDDFT) can achieve the unsupervised recognition of lesion morphology. By introducing the architecture of geometric deep learning and metrics of graph neural networks, gridized density functionals of the fDDFT establish an unsupervised feature-aware mechanism with global convolutional kernels to extract the most likely lesion boundaries and produce lesion segmentation. An AutoEncoder-assisted module reduces the computational complexity from [Formula: see text] to [Formula: see text], thus efficiently speeding up global convolutional operations. We validate their performance utilizing various open-access datasets and discuss limitations. The inference time of each object in large three-dimensional datasets is 1.76 s on average. The proposed gridized density functionals have activation capability synergized with gradient ascent operations, hence can be modularized and embedded in pipelines of modern deep neural networks. Algorithms of geometric stability and similarity convergence also raise the accuracy of unsupervised recognition and segmentation of lesion images. Their performance achieves the standard requirement for conventional deep neural networks; the median dice score is higher than 0.75. The experiment shows that the synergy of fDDFT and a naïve neural network improves the training and inference time by 58% and 51%, respectively, and the dice score raises to 0.9415. This advantage facilitates fast computational modeling in interdisciplinary applications and clinical investigation.

Deep learning for myocardial ischemia auxiliary diagnosis using CZT SPECT myocardial perfusion imaging

BACKGROUND

The World Health Organization reported that cardiovascular disease is the most common cause of death worldwide. On average, one person dies of heart disease every 26 min worldwide. Deep learning approaches are characterized by the appropriate combination of abnormal features based on numerous annotated images. The constructed convolutional neural network (CNN) model can identify normal states of reversible and irreversible myocardial defects and alert physicians for further diagnosis.

METHODS

Cadmium zinc telluride single-photon emission computed tomography myocardial perfusion resting-state images were collected at Chang Gung Memorial Hospital, Kaohsiung Medical Center, Kaohsiung, Taiwan, and were analyzed with a deep learning convolutional neural network to classify myocardial perfusion images for coronary heart diseases.

RESULTS

In these grey-scale images, the heart blood flow distribution was the most crucial feature. The deep learning technique of You Only Look Once was used to determine the myocardial defect area and crop the images. After surrounding noise had been eliminated, a three-dimensional CNN model was used to identify patients with coronary heart diseases. The prediction area under the curve, accuracy, sensitivity, and specificity was 90.97, 87.08, 86.49, and 87.41%, respectively.

CONCLUSION

Our prototype system can considerably reduce the time required for image interpretation and improve the quality of medical care. It can assist clinical experts by offering accurate coronary heart disease diagnosis in practice.

Automatic Screening of Pediatric Renal Ultrasound Abnormalities: Deep Learning and Transfer Learning Approach

BACKGROUND

In recent years, the progress and generalization surrounding portable ultrasonic probes has made ultrasound (US) a useful tool for physicians when making a diagnosis. With the advent of machine learning and deep learning, the development of a computer-aided diagnostic system for screening renal US abnormalities can assist general practitioners in the early detection of pediatric kidney diseases.

OBJECTIVE

In this paper, we sought to evaluate the diagnostic performance of deep learning techniques to classify kidney images as normal and abnormal.

METHODS

We chose 330 normal and 1269 abnormal pediatric renal US images for establishing a model for artificial intelligence. The abnormal images involved stones, cysts, hyperechogenicity, space-occupying lesions, and hydronephrosis. We performed preprocessing of the original images for subsequent deep learning. We redefined the final connecting layers for classification of the extracted features as abnormal or normal from the ResNet-50 pretrained model. The performances of the model were tested by a validation data set using area under the receiver operating characteristic curve, accuracy, specificity, and sensitivity.

RESULTS

The deep learning model, 94 MB parameters in size, based on ResNet-50, was built for classifying normal and abnormal images. The accuracy, (%)/area under curve, of the validated images of stone, cyst, hyperechogenicity, space-occupying lesions, and hydronephrosis were 93.2/0.973, 91.6/0.940, 89.9/0.940, 91.3/0.934, and 94.1/0.996, respectively. The accuracy of normal image classification in the validation data set was 90.1%. Overall accuracy of (%)/area under curve was 92.9/0.959..

CONCLUSIONS

We established a useful, computer-aided model for automatic classification of pediatric renal US images in terms of normal and abnormal categories.

The age factor influencing long-term physical functionality in stroke patients undergoing intra-arterial thrombectomy treatment

The treatment of acute ischemic stroke is heavily time-dependent, and even though, with the most efficient treatment, the long-term functional outcome is still highly variable. In this current study, the authors selected acute ischemic stroke patients who were qualified for intravenous thrombolysis with recombinant tissue plasminogen activator and followed by intra-arterial thrombectomy. With primary outcome defined by the functional level in a 1-year follow-up, we hypothesize that patients with older age are at a disadvantage in post-stroke recovery. However, an age-threshold should be determined to help clinicians in selection of patients to undergo such therapy. This is a retrospective chart review study that include 92 stroke patients in Changhua Christian hospital with a total of 68 evaluation indexes recorded. The current study utilized the forward stepwise regression model whose Adj-R2 and P value in search of important variables for outcome prediction. The chngpt package in R indicated the threshold point of the age factor directing the better future functionality of the stroke patients. Datasets revealed the threshold of the age set at 79 the most appropriate. Admission Barthel Index, Age, ipsilateral internal carotid artery resistance index (ICA RI), ipsilateral vertebral artery (VA) PI, contralateral middle cerebral artery (MCA) stenosis, contralateral external carotid artery (ECA) RI, and in-hospital pneumonia are the significant predicting variables. The higher the age, in-hospital pneumonia, contralateral MCA stenosis, ipsilateral ICA RI and ipsilateral VA PI, the less likely patient to recover from functional deficits as the result of acute ischemic stroke; the higher the value of contralateral ECA RI and admission Barthel Index, the better chance to full functional recovery at 1-year follow up. Parameters of pre-intervention datasets could provide important information to aid first-line clinicians in decision making. Especially, in patients whose age is above 79 receives diminish return in the benefit to undergo such intervention and should be considered seriously by both the patients and the physicians.

Ensemble classification and segmentation for intracranial metastatic tumors on MRI images based on 2D U-nets

The extraction of brain tumor tissues in 3D Brain Magnetic Resonance Imaging (MRI) plays an important role in diagnosis before the gamma knife radiosurgery (GKRS). In this article, the post-contrast T1 whole-brain MRI images had been collected by Taipei Veterans General Hospital (TVGH) and stored in DICOM format (dated from 1999 to 2018). The proposed method starts with the active contour model to get the region of interest (ROI) automatically and enhance the image contrast. The segmentation models are trained by MRI images with tumors to avoid imbalanced data problem under model construction. In order to achieve this objective, a two-step ensemble approach is used to establish such diagnosis, first, classify whether there is any tumor in the image, and second, segment the intracranial metastatic tumors by ensemble neural networks based on 2D U-Net architecture. The ensemble for classification and segmentation simultaneously also improves segmentation accuracy. The result of classification achieves a F1-measure of [Formula: see text], while the result of segmentation achieves an IoU of [Formula: see text] and a DICE score of [Formula: see text]. Significantly reduce the time for manual labeling from 30 min to 18 s per patient.

Can 3D artificial intelligence models outshine 2D ones in the detection of intracranial metastatic tumors on magnetic resonance images?

BACKGROUND

This study aimed to compare the prediction performance of two-dimensional (2D) and three-dimensional (3D) semantic segmentation models for intracranial metastatic tumors with a volume ≥ 0.3 mL.

METHODS

We used postcontrast T1 whole-brain magnetic resonance (MR), which was collected from Taipei Veterans General Hospital (TVGH). Also, the study was approved by the institutional review board (IRB) of TVGH. The 2D image segmentation model does not fully use the spatial information between neighboring slices, whereas the 3D segmentation model does. We treated the U-Net as the basic model for 2D and 3D architectures.

RESULTS

For the prediction of intracranial metastatic tumors, the area under the curve (AUC) of the 3D model was 87.6% and that of the 2D model was 81.5%.

CONCLUSION

Building a semantic segmentation model based on 3D deep convolutional neural networks might be crucial to achieve a high detection rate in clinical applications for intracranial metastatic tumors.

Can a Deep-learning Model for the Automated Detection of Vertebral Fractures Approach the Performance Level of Human Subspecialists?

BACKGROUND

Vertebral fractures are the most common osteoporotic fractures in older individuals. Recent studies suggest that the performance of artificial intelligence is equal to humans in detecting osteoporotic fractures, such as fractures of the hip, distal radius, and proximal humerus. However, whether artificial intelligence performs as well in the detection of vertebral fractures on plain lateral spine radiographs has not yet been reported.

QUESTIONS/PURPOSES

(1) What is the accuracy, sensitivity, specificity, and interobserver reliability (kappa value) of an artificial intelligence model in detecting vertebral fractures, based on Genant fracture grades, using plain lateral spine radiographs compared with values obtained by human observers? (2) Do patients' clinical data, including the anatomic location of the fracture (thoracic or lumbar spine), T-score on dual-energy x-ray absorptiometry, or fracture grade severity, affect the performance of an artificial intelligence model? (3) How does the artificial intelligence model perform on external validation?

METHODS

Between 2016 and 2018, 1019 patients older than 60 years were treated for vertebral fractures in our institution. Seventy-eight patients were excluded because of missing CT or MRI scans (24% [19]), poor image quality in plain lateral radiographs of spines (54% [42]), multiple myeloma (5% [4]), and prior spine instrumentation (17% [13]). The plain lateral radiographs of 941 patients (one radiograph per person), with a mean age of 76 ± 12 years, and 1101 vertebral fractures between T7 and L5 were retrospectively evaluated for training (n = 565), validating (n = 188), and testing (n = 188) of an artificial intelligence deep-learning model. The gold standard for diagnosis (ground truth) of a vertebral fracture is the interpretation of the CT or MRI reports by a spine surgeon and a radiologist independently. If there were any disagreements between human observers, the corresponding CT or MRI images would be rechecked by them together to reach a consensus. For the Genant classification, the injured vertebral body height was measured in the anterior, middle, and posterior third. Fractures were classified as Grade 1 (< 25%), Grade 2 (26% to 40%), or Grade 3 (> 40%). The framework of the artificial intelligence deep-learning model included object detection, data preprocessing of radiographs, and classification to detect vertebral fractures. Approximately 90 seconds was needed to complete the procedure and obtain the artificial intelligence model results when applied clinically. The accuracy, sensitivity, specificity, interobserver reliability (kappa value), receiver operating characteristic curve, and area under the curve (AUC) were analyzed. The bootstrapping method was applied to our testing dataset and external validation dataset. The accuracy, sensitivity, and specificity were used to investigate whether fracture anatomic location or T-score in dual-energy x-ray absorptiometry report affected the performance of the artificial intelligence model. The receiver operating characteristic curve and AUC were used to investigate the relationship between the performance of the artificial intelligence model and fracture grade. External validation with a similar age population and plain lateral radiographs from another medical institute was also performed to investigate the performance of the artificial intelligence model.

RESULTS

The artificial intelligence model with ensemble method demonstrated excellent accuracy (93% [773 of 830] of vertebrae), sensitivity (91% [129 of 141]), and specificity (93% [644 of 689]) for detecting vertebral fractures of the lumbar spine. The interobserver reliability (kappa value) of the artificial intelligence performance and human observers for thoracic and lumbar vertebrae were 0.72 (95% CI 0.65 to 0.80; p < 0.001) and 0.77 (95% CI 0.72 to 0.83; p < 0.001), respectively. The AUCs for Grades 1, 2, and 3 vertebral fractures were 0.919, 0.989, and 0.990, respectively. The artificial intelligence model with ensemble method demonstrated poorer performance for discriminating normal osteoporotic lumbar vertebrae, with a specificity of 91% (260 of 285) compared with nonosteoporotic lumbar vertebrae, with a specificity of 95% (222 of 234). There was a higher sensitivity 97% (60 of 62) for detecting osteoporotic (dual-energy x-ray absorptiometry T-score ≤ -2.5) lumbar vertebral fractures, implying easier detection, than for nonosteoporotic vertebral fractures (83% [39 of 47]). The artificial intelligence model also demonstrated better detection of lumbar vertebral fractures compared with detection of thoracic vertebral fractures based on the external dataset using various radiographic techniques. Based on the dataset for external validation, the overall accuracy, sensitivity, and specificity on bootstrapping method were 89%, 83%, and 95%, respectively.

CONCLUSION

The artificial intelligence model detected vertebral fractures on plain lateral radiographs with high accuracy, sensitivity, and specificity, especially for osteoporotic lumbar vertebral fractures (Genant Grades 2 and 3). The rapid reporting of results using this artificial intelligence model may improve the efficiency of diagnosing vertebral fractures. The testing model is available at http://140.113.114.104/vght_demo/corr/. One or multiple plain lateral radiographs of the spine in the Digital Imaging and Communications in Medicine format can be uploaded to see the performance of the artificial intelligence model.

LEVEL OF EVIDENCE

Level II, diagnostic study.

Accurate diagnosis of endoscopic mucosal healing in ulcerative colitis using deep learning and machine learning

BACKGROUND

In clinical applications, mucosal healing is a therapeutic goal in patients with ulcerative colitis (UC). Endoscopic remission is associated with lower rates of colectomy, relapse, hospitalization, and colorectal cancer. Differentiation of mucosal inflammatory status depends on the experience and subjective judgments of clinical physicians. We developed a computer-aided diagnostic system using deep learning and machine learning (DLML-CAD) to accurately diagnose mucosal healing in UC patients.

METHODS

We selected 856 endoscopic colon images from 54 UC patients (643 images with endoscopic score 0-1 and 213 with score 2-3) from the endoscopic image database at Tri-Service General Hospital, Taiwan. Endoscopic grading using the Mayo endoscopic subscore (MES 0-3) was performed by two reviewers. A pretrained neural network extracted image features, which were used to train three different classifiers-deep neural network (DNN), support vector machine (SVM), and k-nearest neighbor (k-NN) network.

RESULTS

DNN classified MES 0 to 1, representing mucosal healing, vs MES 2 to 3 images with 93.8% accuracy (sensitivity 84.6%, specificity 96.9%); SVM had 94.1% accuracy (sensitivity 89.2%, specificity 95.8%); and k-NN had 93.4% accuracy (sensitivity 86.2%, specificity 95.8%). Combined, ensemble learning achieved 94.5% accuracy (sensitivity 89.2%, specificity 96.3%). The system further differentiated between MES 0, representing complete mucosal healing, and MES 1 images with 89.1% accuracy (sensitivity 82.3%, specificity 92.2%).

CONCLUSION

Our DLML-CAD diagnosis achieved 94.5% accuracy for endoscopic mucosal healing and 89.0% accuracy for complete mucosal healing. This system can provide clinical physicians with an accurate auxiliary diagnosis in treating UC.

Deep learning and ensemble stacking technique for differentiating polypoidal choroidal vasculopathy from neovascular age-related macular degeneration

Polypoidal choroidal vasculopathy (PCV) and neovascular age-related macular degeneration (nAMD) share some similarity in clinical imaging manifestations. However, their disease entity and treatment strategy as well as visual outcomes are very different. To distinguish these two vision-threatening diseases is somewhat challenging but necessary. In this study, we propose a new artificial intelligence model using an ensemble stacking technique, which combines a color fundus photograph-based deep learning (DL) model and optical coherence tomography-based biomarkers, for differentiation of PCV from nAMD. Furthermore, we introduced multiple correspondence analysis, a method of transforming categorical data into principal components, to handle the dichotomous data for combining with another image DL system. This model achieved a robust performance with an accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve of 83.67%, 80.76%, 84.72%, and 88.57%, respectively, by training nearly 700 active cases with suitable imaging quality and transfer learning architecture. This work could offer an alternative method of developing a multimodal DL model, improve its efficiency for distinguishing different diseases, and facilitate the broad application of medical engineering in a DL model design.

The Clinical Application of the Deep Learning Technique for Predicting Trigger Origins in Patients With Paroxysmal Atrial Fibrillation With Catheter Ablation

Background:

Non–pulmonary vein (NPV) trigger has been reported as an important predictor of recurrence post–atrial fibrillation ablation. Elimination of NPV triggers can reduce the recurrence of postablation atrial fibrillation. Deep learning was applied to preablation pulmonary vein computed tomography geometric slices to create a prediction model for NPV triggers in patients with paroxysmal atrial fibrillation.

Methods:

We retrospectively analyzed 521 patients with paroxysmal atrial fibrillation who underwent catheter ablation of paroxysmal atrial fibrillation. Among them, pulmonary vein computed tomography geometric slices from 358 patients with nonrecurrent atrial fibrillation (1–3 mm interspace per slice, 20–200 slices for each patient, ranging from the upper border of the left atrium to the bottom of the heart, for a total of 23 683 images of slices) were used in the deep learning process, the ResNet34 of the neural network, to create the prediction model of the NPV trigger. There were 298 (83.2%) patients with only pulmonary vein triggers and 60 (16.8%) patients with NPV triggers±pulmonary vein triggers. The patients were randomly assigned to either training, validation, or test groups, and their data were allocated according to those sets. The image datasets were split into training (n=17 340), validation (n=3491), and testing (n=2852) groups, which had completely independent sets of patients.

Results:

The accuracy of prediction in each pulmonary vein computed tomography image for NPV trigger was up to 82.4±2.0%. The sensitivity and specificity were 64.3±5.4% and 88.4±1.9%, respectively. For each patient, the accuracy of prediction for a NPV trigger was 88.6±2.3%. The sensitivity and specificity were 75.0±5.8% and 95.7±1.8%, respectively. The area under the curve for each image and patient were 0.82±0.01 and 0.88±0.07, respectively.

Conclusions:

The deep learning model using preablation pulmonary vein computed tomography can be applied to predict the trigger origins in patients with paroxysmal atrial fibrillation receiving catheter ablation. The application of this model may identify patients with a high risk of NPV trigger before ablation.

Comparisons of traditional electrocardiographic criteria for left and right ventricular hypertrophy in young Asian women: The CHIEF heart study

The performance of electrocardiographic (ECG) voltage criteria to identify left and right ventricular hypertrophy (LVH and RVH) in young Asian female adults have not been clarified so far.In a sample of 255 military young female adults, aged 25.2 years on average, echocardiographic LVH was respectively defined as the left ventricular mass (LVM) indexed by body surface area (BSA) (≥88 g/m) and by height (≥41 g/m), and RVH was defined as anterior right ventricular wall thickness >5.2 mm. The performance of ECG voltage criteria for the echocardiographic LVH and RVH were assessed by area under curve (AUC) of receiver operating characteristic (ROC) curve to estimate sensitivity and specificity.For the Sokolow-Lyon (the maximum of SV1 or SV2 + RV5 or RV6) and Cornell (RaVL + SV3) voltage criteria with the LVM/BSA ≥88 g/m, the AUC of ROC curves were 0.66 (95% confidence intervals [CI]: 0.52-0.81, P = .039) and 0.61 (95% CI: 0.44-0.77, P = .18), respectively. For these 2 ECG voltage criteria with the LVM/height ≥41 g/m, the AUC of ROC curves were 0.64 (95% CI: 0.52-0.75, P = 0.11) and 0.73 (95% CI: 0.61-0.85, P = 0.0074), respectively. The best cut-off points selected for the Sokolow-Lyon and Cornell voltage criteria with echocardiographic LVH in young Asian females were 26 mm and 6 mm, respectively. In contrast, all the AUC of ROC curves were less than 0.60 and not significant according to the Sokolow-Lyon (the maximum of RV1 + SV5 or V6) and Myers' voltage criteria (eg, the voltage of R wave in V1 and the ratios of R/S in V1, V5 and V6) with echocardiographic RVH.There was a suggestion that the ECG voltage criteria to screen the presence of LVH should be adjusted for the young Asian female adults, and with regard to RVH, the ECG voltage criteria were found ineffective.

A 12-Lead ECG-Based System With Physiological Parameters and Machine Learning to Identify Right Ventricular Hypertrophy in Young Adults

OBJECTIVE

The presence of right ventricular hypertrophy (RVH) accounts for approximately 5-10% in young adults. The sensitivity estimated by commonly used 12-lead electrocardiographic (ECG) criteria for identifying the presence of RVH is under 20% in the general population. The aim of this study is to develop a 12-lead ECG system with the related information of age, body height and body weight via machine learning to increase the sensitivity and the precision for detecting RVH.

METHOD

In a sample of 1,701 males, aged 17-45 years, support vector machine is used for the training of 31 parameters including age, body height and body weight in addition to 28 ECG data such as axes, intervals and wave voltages as the inputs to link the output RVH. The RVH is defined on the echocardiographic finding for young males as right ventricular anterior wall thickness > 5.5 mm.

RESULTS

On the system goal for increasing sensitivity, the specificity is controlled around 70-75% and all data tested in the proposed method show competent sensitivity up to 70.3%. The values of area under curve of receiver operating characteristic curve and precision-recall curve using the proposed method are 0.780 and 0.285, respectively, which are better than 0.518 and 0.112 using the Sokolow-Lyon voltage criterion, respectively, for detecting unspecific RVH.

CONCLUSION

We present a method using simple physiological parameters with ECG data to effectively identify more than 70% of the RVH among young adults. Clinical Impact: This system provides a fast, precise and feasible diagnosis tool to screen RVH.

Boolean function network analysis of time course liver transcriptome data to reveal novel circadian transcriptional regulators in mammals

BACKGROUND

Many biological processes in mammals are subject to circadian control at the molecular level. Disruption of circadian rhythms has been demonstrated to be associated with a wide range of diseases, such as diabetes mellitus, mental disorders, and cancer. Although the core circadian genes are well established, there are multiple reports of novel peripheral circadian regulators. The goal of this study was to provide a comprehensive computational analysis to identify novel potential circadian transcriptional regulators.

METHODS

To fulfill the aforementioned goal, we applied a Boolean function network method to analyze the microarray time course mouse and rat liver datasets available in the literature. The inferred direct pairwise relations were further investigated using the functional annotation tool. This approach generated a list of transcription factors (TFs) and cofactors, which were associated with significantly enriched circadian gene ontology (GO) categories.

RESULTS

As a result, we identified 93 transcriptional circadian regulators in mouse and 95 transcriptional circadian regulators in rat. Of these, 19 regulators in mouse and 21 regulators in rat were known, whereas the rest were novel. Furthermore, we validated novel circadian TFs with bioinformatics databases, previous large-scale circadian studies, and related small-scale studies. Moreover, according to predictions inferred from ChIP-Seq experiments reported in the database, 40 of our candidate circadian regulators were confirmed to have circadian genes as direct regulatory targets. In addition, we annotated candidate circadian regulators with disorders that were often associated with disruptions of circadian rhythm in the literature.

CONCLUSION

In summary, our computational analysis, which was followed by an extensive verification by means of a literature review, can contribute to translational study from endocrinology to cancer research and provide insights for future investigation.

Correction: Exploring candidate biological functions by Boolean Function Networks for Saccharomyces cerevisiae

[This corrects the article DOI: 10.1371/journal.pone.0185475.].

Reversed ophthalmic artery flow following ischemic stroke: a possible predictor of outcomes following carotid artery stenting

Stroke is the leading cause of death worldwide and stenosis of the carotid artery accounts for more than half of all cases. Carotid duplex is an effective non-invasive ultrasound test which identifies stroke patients with moderate to severe carotid stenosis who are candidates for preventative intervention to reduce the risk of recurrence. In patients with moderate to severe carotid stenosis, reversed ophthalmic artery flow (ROAF) is often observed at the time of the carotid duplex scan. In this study, we investigated whether ROAF, denoting exhaustion of cerebral collateral flow in ischemic stroke patients affected mid-term functional outcomes following carotid artery stenting (CAS) procedures. In total, 144 consecutive patients with a first episode of ischemic stroke and subsequent CAS procedure conducted between January 2010 and November 2014 at Changhua Christian Hospital, Taiwan were included. Clinical data were obtained by medical record review. Disability was assessed at two time points by utilising the Barthel Index (BI) and modified Rankin Scale (mRS) before CAS and 12 months post-CAS. Among 85 patients presenting without ROAF, 48/85 (56.4%) had improved mRS scores following stenting. The condition remained unchanged (stationary) in 36/85 (43.5%) patients after stenting and one patient exhibited deteriorated condition 1/85(1.1%). In contrast, among the 59 patients presenting with ROAF, 24/59 (40.6%) had improved mRS score following stenting. The condition remained unchanged (stationary) in the remaining 35/59 (59.3%) patients after stenting, and no patient exhibited deteriorated condition 0/59 (0 %). This study provides evidence that CAS is a valid and effective treatment option regardless of whether patients exhibited ROAF or not. Patients without ROAF were significantly more likely to have improved mid-term functional outcomes compared to those with ROAF. In the group without ROAF admission, CRP may play a role in predicting subsequent functional outcomes, whereas admission Barthel Index was a predictor of outcome in the ROAF group.

A gene profiling deconvolution approach to estimating immune cell composition from complex tissues

Background

A new emerged cancer treatment utilizes intrinsic immune surveillance mechanism that is silenced by those malicious cells. Hence, studies of tumor infiltrating lymphocyte populations (TILs) are key to the success of advanced treatments. In addition to laboratory methods such as immunohistochemistry and flow cytometry, in silico gene expression deconvolution methods are available for analyses of relative proportions of immune cell types.

Results

Herein, we used microarray data from the public domain to profile gene expression pattern of twenty-two immune cell types. Initially, outliers were detected based on the consistency of gene profiling clustering results and the original cell phenotype notation. Subsequently, we filtered out genes that are expressed in non-hematopoietic normal tissues and cancer cells. For every pair of immune cell types, we ran t-tests for each gene, and defined differentially expressed genes (DEGs) from this comparison. Equal numbers of DEGs were then collected as candidate lists and numbers of conditions and minimal values for building signature matrixes were calculated. Finally, we used v -Support Vector Regression to construct a deconvolution model. The performance of our system was finally evaluated using blood biopsies from 20 adults, in which 9 immune cell types were identified using flow cytometry. The present computations performed better than current state-of-the-art deconvolution methods.

Conclusions

Finally, we implemented the proposed method into R and tested extensibility and usability on Windows, MacOS, and Linux operating systems. The method, MySort, is wrapped as the Galaxy platform pluggable tool and usage details are available at https://testtoolshed.g2.bx.psu.edu/view/moneycat/mysort/e3afe097e80a.

Accurate Classification of Diminutive Colorectal Polyps Using Computer-Aided Analysis

BACKGROUND & AIMS

Narrow-band imaging is an image-enhanced form of endoscopy used to observed microstructures and capillaries of the mucosal epithelium which allows for real-time prediction of histologic features of colorectal polyps. However, narrow-band imaging expertise is required to differentiate hyperplastic from neoplastic polyps with high levels of accuracy. We developed and tested a system of computer-aided diagnosis with a deep neural network (DNN-CAD) to analyze narrow-band images of diminutive colorectal polyps.

METHODS

We collected 1476 images of neoplastic polyps and 681 images of hyperplastic polyps, obtained from the picture archiving and communications system database in a tertiary hospital in Taiwan. Histologic findings from the polyps were also collected and used as the reference standard. The images and data were used to train the DNN. A test set of images (96 hyperplastic and 188 neoplastic polyps, smaller than 5 mm), obtained from patients who underwent colonoscopies from March 2017 through August 2017, was then used to test the diagnostic ability of the DNN-CAD vs endoscopists (2 expert and 4 novice), who were asked to classify the images of the test set as neoplastic or hyperplastic. Their classifications were compared with findings from histologic analysis. The primary outcome measures were diagnostic accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and diagnostic time. The accuracy, sensitivity, specificity, PPV, NPV, and diagnostic time were compared among DNN-CAD, the novice endoscopists, and the expert endoscopists. The study was designed to detect a difference of 10% in accuracy by a 2-sided McNemar test.

RESULTS

In the test set, the DNN-CAD identified neoplastic or hyperplastic polyps with 96.3% sensitivity, 78.1% specificity, a PPV of 89.6%, and a NPV of 91.5%. Fewer than half of the novice endoscopists classified polyps with a NPV of 90% (their NPVs ranged from 73.9% to 84.0%). DNN-CAD classified polyps as neoplastic or hyperplastic in 0.45 ± 0.07 seconds-shorter than the time required by experts (1.54 ± 1.30 seconds) and nonexperts (1.77 ± 1.37 seconds) (both P < .001). DNN-CAD classified polyps with perfect intra-observer agreement (kappa score of 1). There was a low level of intra-observer and inter-observer agreement in classification among endoscopists.

CONCLUSIONS

We developed a system called DNN-CAD to identify neoplastic or hyperplastic colorectal polyps less than 5 mm. The system classified polyps with a PPV of 89.6%, and a NPV of 91.5%, and in a shorter time than endoscopists. This deep-learning model has potential for not only endoscopic image recognition but for other forms of medical image analysis, including sonography, computed tomography, and magnetic resonance images.

Exploring candidate biological functions by Boolean Function Networks for Saccharomyces cerevisiae

The great amount of gene expression data has brought a big challenge for the discovery of Gene Regulatory Network (GRN). For network reconstruction and the investigation of regulatory relations, it is desirable to ensure directness of links between genes on a map, infer their directionality and explore candidate biological functions from high-throughput transcriptomic data. To address these problems, we introduce a Boolean Function Network (BFN) model based on techniques of hidden Markov model (HMM), likelihood ratio test and Boolean logic functions. BFN consists of two consecutive tests to establish links between pairs of genes and check their directness. We evaluate the performance of BFN through the application to S. cerevisiae time course data. BFN produces regulatory relations which show consistency with succession of cell cycle phases. Furthermore, it also improves sensitivity and specificity when compared with alternative methods of genetic network reverse engineering. Moreover, we demonstrate that BFN can provide proper resolution for GO enrichment of gene sets. Finally, the Boolean functions discovered by BFN can provide useful insights for the identification of control mechanisms of regulatory processes, which is the special advantage of the proposed approach. In combination with low computational complexity, BFN can serve as an efficient screening tool to reconstruct genes relations on the whole genome level. In addition, the BFN approach is also feasible to a wide range of time course datasets.

Clonal spread of carbapenem-resistant Acinetobacter baumannii across a community hospital and its affiliated long-term care facilities: A cross sectional study

BACKGROUND

The global spread of carbapenem-resistant Acinetobacter baumannii (CRAB) is now a public health problem. In Taiwan, the relationship of the CRAB circulation between long-term care facilities (LTCFs) and acute care hospitals remains unclear. Here, we use molecular epidemiologic methods to describe the transmission of CRAB isolates between a community hospital and its affiliated LTCFs.

METHODS

Subjects localized in eight LTCFs who were not admitted acute care hospitals in recent a year were enrolled in this study. CRAB isolates were collected during June 1, 2015 and December 31, 2015. DNA fingerprinting was performed by repetitive extragenic palindromic sequence-based polymerase chain reaction (Rep-PCR) and multilocus sequence typing (MLST). Multiplex-PCR amplification for the detection of bla_OXA genes and beta-lactamase genes was performed.

RESULTS

Twenty one subjects were enrolled. The major hospital admission diagnoses among the 21 subjects were pneumonia (71.4%). Genotyping of CRAB isolates by Rep-PCR revealed that a major clone, designated as type III, comprised fifteen of 21 (71.4%) isolates taken from 5 LTCFs and one study hospital. The isolates with type III were subtyped by PubMLST into 4 ST types. The most prevalent bla_OXA genes in these isolates were bla_OXA-23-like (85.70%, 18/21). Twenty isolates carried bla_SHV. CONCLUSION: Clonal spread of bla_OxA-23-carrying CRABs was found around LTCFs and the affiliated hospital. In Taiwan, it is important for the government to focus attention on the importance of identifying and tracing CRAB infections in LTCFs.

Is carotid sonography a useful tool for predicting functional capabilities in ischemic stroke patients following carotid artery stenting?

Carotid stenosis is a major cause of stroke and timely intervention with stenting manipulation can significantly reduce the risk of secondary stroke. The impact of stenting procedures on patient functional capabilities has not yet been explored. The primary aim of this study was to examine associations between periprocedural carotid sonography parameters and post-treatment functional capabilities in stroke patients.Sixty-seven patients who received carotid stenting at 1 angiography laboratory were included. Prestenting and poststenting carotid duplex data were recorded and resistance index (RI) differences at various carotid system locations were compared. The modified Rankin Scale (mRS) was used to assess functional capability. All of the studied parameters were analyzed by SPSS (version 16.0, SPSS Inc, Chicago, IL).Following stenting, mRS scores improved (n = 44) or remained stationary (n = 23). Net contralateral internal carotid artery (ICA) RI for patients with improved mRS was lower compared to that for patients with stationary mRS (median = 0.040 vs 0.11; P = 0.003). The contralateral common carotid artery RI before and after stenting differed significantly (P < 0.050) in both. The ipsilateral ICA RI differed (P < 0.050) only in patients with improved mRS. The difference in mean transit time, Barthel index, net ipsilateral ICA RI, net contralateral external carotid artery RI, postipsilateral common carotid artery RI, and postipsilateral ICA RI differed significantly between different baseline stroke severity groups (P < 0.050).Carotid artery stenting improved physical function in a proportion of ischemic stroke patients with carotid stenosis. Carotid ultrasound is a useful assessment tool to predict likely functional outcomes following carotid artery stenting.

A Sliced Inverse Regression (SIR) Decoding the Forelimb Movement from Neuronal Spikes in the Rat Motor Cortex

Several neural decoding algorithms have successfully converted brain signals into commands to control a computer cursor and prosthetic devices. A majority of decoding methods, such as population vector algorithms (PVA), optimal linear estimators (OLE), and neural networks (NN), are effective in predicting movement kinematics, including movement direction, speed and trajectory but usually require a large number of neurons to achieve desirable performance. This study proposed a novel decoding algorithm even with signals obtained from a smaller numbers of neurons. We adopted sliced inverse regression (SIR) to predict forelimb movement from single-unit activities recorded in the rat primary motor (M1) cortex in a water-reward lever-pressing task. SIR performed weighted principal component analysis (PCA) to achieve effective dimension reduction for nonlinear regression. To demonstrate the decoding performance, SIR was compared to PVA, OLE, and NN. Furthermore, PCA and sequential feature selection (SFS) which are popular feature selection techniques were implemented for comparison of feature selection effectiveness. Among SIR, PVA, OLE, PCA, SFS, and NN decoding methods, the trajectories predicted by SIR (with a root mean square error, RMSE, of 8.47 ± 1.32 mm) was closer to the actual trajectories compared with those predicted by PVA (30.41 ± 11.73 mm), OLE (20.17 ± 6.43 mm), PCA (19.13 ± 0.75 mm), SFS (22.75 ± 2.01 mm), and NN (16.75 ± 2.02 mm). The superiority of SIR was most obvious when the sample size of neurons was small. We concluded that SIR sorted the input data to obtain the effective transform matrices for movement prediction, making it a robust decoding method for conditions with sparse neuronal information.

Dynamic change of surface microbiota with different environmental cleaning methods between two wards in a hospital

Terminal disinfection and daily cleaning have been performed in hospitals in Taiwan for many years to reduce the risks of healthcare-associated infections. However, the effectiveness of these cleaning approaches and dynamic changes of surface microbiota upon cleaning remain unclear. Here, we report the surface changes of bacterial communities with terminal disinfection and daily cleaning in a medical intensive care unit (MICU) and only terminal disinfection in a respiratory care center (RCC) using 16s ribosomal RNA (rRNA) metagenomics. A total of 36 samples, including 9 samples per sampling time, from each ward were analysed. The clinical isolates were recorded during the sampling time. A large amount of microbial diversity was detected, and human skin microbiota (HSM) was predominant in both wards. In addition, the colonization rate of the HSM in the MICU was higher than that in the RCC, especially for Moraxellaceae. A higher alpha-diversity (p = 0.005519) and a lower UniFrac distance was shown in the RCC due to the lack of daily cleaning. Moreover, a significantly higher abundance among Acinetobacter sp., Streptococcus sp. and Pseudomonas sp. was shown in the RCC compared to the MICU using the paired t test. We concluded that cleaning changes might contribute to the difference in diversity between two wards.

Phylogenetic tree construction using trinucleotide usage profile (TUP)

BACKGROUND

It has been a challenging task to build a genome-wide phylogenetic tree for a large group of species containing a large number of genes with long nucleotides sequences. The most popular method, called feature frequency profile (FFP-k), finds the frequency distribution for all words of certain length k over the whole genome sequence using (overlapping) windows of the same length. For a satisfactory result, the recommended word length (k) ranges from 6 to 15 and it may not be a multiple of 3 (codon length). The total number of possible words needed for FFP-k can range from 4⁶=4096 to 4¹⁵.

RESULTS

We propose a simple improvement over the popular FFP method using only a typical word length of 3. A new method, called Trinucleotide Usage Profile (TUP), is proposed based only on the (relative) frequency distribution using non-overlapping windows of length 3. The total number of possible words needed for TUP is 4³=64, which is much less than the total count for the recommended optimal "resolution" for FFP. To build a phylogenetic tree, we propose first representing each of the species by a TUP vector and then using an appropriate distance measure between pairs of the TUP vectors for the tree construction. In particular, we propose summarizing a DNA sequence by a matrix of three rows corresponding to three reading frames, recording the frequency distribution of the non-overlapping words of length 3 in each of the reading frame. We also provide a numerical measure for comparing trees constructed with various methods.

CONCLUSIONS

Compared to the FFP method, our empirical study showed that the proposed TUP method is more capable of building phylogenetic trees with a stronger biological support. We further provide some justifications on this from the information theory viewpoint. Unlike the FFP method, the TUP method takes the advantage that the starting of the first reading frame is (usually) known. Without this information, the FFP method could only rely on the frequency distribution of overlapping words, which is the average (or mixture) of the frequency distributions of three possible reading frames. Consequently, we show (from the entropy viewpoint) that the FFP procedure could dilute important gene information and therefore provides less accurate classification.

First-ever ischemic stroke in elderly patients: predictors of functional outcome following carotid artery stenting

Age is an important risk factor for stroke, and carotid artery stenosis is the primary cause of first-ever ischemic stroke. Timely intervention with stenting procedures can effectively prevent secondary stroke; however, the impact of stenting on various periprocedural physical functionalities has never been thoroughly investigated. The primary aim of this study was to investigate whether prestenting characteristics were associated with long-term functional outcomes in patients presenting with first-ever ischemic stroke. The secondary aim was to investigate whether patient age was an important factor in outcomes following stenting, measured by the modified Rankin scale (mRS). In total, 144 consecutive patients with first-ever ischemic stroke who underwent carotid artery stenting from January 2010 to November 2014 were included. Clinical data were obtained by review of medical records. The Barthel index (BI) and mRS were used to assess disability before stenting and at 12-month follow-up. In total, 72/144 patients showed improvement (mRS[+]), 71 showed stationary and one showed deterioration in condition (mRS[-]). The prestenting parameters, ratio of cerebral blood volume (1.41 vs 1.2 for mRS[-] vs mRS[+]), BI (75 vs 85), and high-sensitivity C-reactive protein (hsCRP 5.0 vs 3.99), differed significantly between the two outcome groups (P<0.05). The internal carotid artery/common carotid artery ratio (P=0.011), BI (P=0.019), ipsilateral internal carotid artery resistance index (P=0.003), and HbA1c (P=0.039) were all factors significantly associated with patient age group. There was no significant association between age and poststenting outcome measured by mRS with 57% of patients in the ≥75 years age group showing mRS(-) and 43% showing mRS(+) (P=0.371). Our findings indicate that in our elderly patient series, carotid artery stenting may benefit a significant proportion of carotid stenotic patients regardless of age. Ratio of cerebral blood volume, BI, and admission hsCRP could serve as important predictors of mRS improvement and may facilitate differentiation of patients at baseline.

Impact of baseline characteristics on outcomes of carotid artery stenting in acute ischemic stroke patients

Carotid artery stenting is an effective treatment for ischemic stroke patients with moderate-to-severe carotid artery stenosis. However, the midterm outcome for patients undergoing this procedure varies considerably with baseline characteristics. To determine the impact of baseline characteristics on outcomes following carotid artery stenting, data from 107 eligible patients with a first episode of ischemic stroke were collected by retrospective chart review. A modified Rankin Scale (mRS) was used to divide patients into two baseline groups, mRS ≤2 and mRS >2. A three-step decision-tree statistical analysis was conducted. After weighting the decision-tree parameters, the following impact hierarchy was obtained: admission low-density lipoprotein, gouty arthritis, chronic kidney disease, ipsilateral common carotid artery resistance index, contralateral ophthalmic artery resistance index, sex, and dyslipidemia. The finite-state machine model demonstrated that, in patients with baseline mRS ≤2, 46% had an improved mRS score at follow-up, whereas 54% had a stable mRS score. In patients with baseline mRS >2, a stable mRS score was observed in 75%, improved score in 23%, and a poorer score in 2%. Admission low-density lipoprotein was the strongest predictive factor influencing poststenting outcome. In addition, our study provides further evidence that carotid artery stenting can be of benefit in first-time ischemic stroke patients with baseline mRS scores >2.

Sufficient dimension reduction with additional information

Sufficient dimension reduction is widely applied to help model building between the response [Formula: see text] and covariate [Formula: see text] In some situations, we also collect additional covariate [Formula: see text] that has better performance in predicting [Formula: see text], but has a higher obtaining cost, than [Formula: see text] While constructing a predictive model for [Formula: see text] based on [Formula: see text] is straightforward, this strategy is not applicable since [Formula: see text] is not available for future observations in which the constructed model is to be applied. As a result, the aim of the study is to build a predictive model for [Formula: see text] based on [Formula: see text] only, where the available data is [Formula: see text] A naive method is to conduct analysis using [Formula: see text] directly, but ignoring [Formula: see text] can cause the problem of inefficiency. On the other hand, it is not trivial to utilize the information of [Formula: see text] to infer [Formula: see text], either. In this article, we propose a two-stage dimension reduction method for [Formula: see text] that is able to utilize the information of [Formula: see text] In the breast cancer data, the risk score constructed from the two-stage method can well separate patients with different survival experiences. In the Pima data, the two-stage method requires fewer components to infer the diabetes status, while achieving higher classification accuracy than the conventional method.

Inference of biological pathway from gene expression profiles by time delay boolean networks

One great challenge of genomic research is to efficiently and accurately identify complex gene regulatory networks. The development of high-throughput technologies provides numerous experimental data such as DNA sequences, protein sequence, and RNA expression profiles makes it possible to study interactions and regulations among genes or other substance in an organism. However, it is crucial to make inference of genetic regulatory networks from gene expression profiles and protein interaction data for systems biology. This study will develop a new approach to reconstruct time delay boolean networks as a tool for exploring biological pathways. In the inference strategy, we will compare all pairs of input genes in those basic relationships by their corresponding p-scores for every output gene. Then, we will combine those consistent relationships to reveal the most probable relationship and reconstruct the genetic network. Specifically, we will prove that O(log n) state transition pairs are sufficient and necessary to reconstruct the time delay boolean network of n nodes with high accuracy if the number of input genes to each gene is bounded. We also have implemented this method on simulated and empirical yeast gene expression data sets. The test results show that this proposed method is extensible for realistic networks.

Evolution of domain architectures and catalytic functions of enzymes in metabolic systems

Domain architectures and catalytic functions of enzymes constitute the centerpieces of a metabolic network. These types of information are formulated as a two-layered network consisting of domains, proteins, and reactions-a domain-protein-reaction (DPR) network. We propose an algorithm to reconstruct the evolutionary history of DPR networks across multiple species and categorize the mechanisms of metabolic systems evolution in terms of network changes. The reconstructed history reveals distinct patterns of evolutionary mechanisms between prokaryotic and eukaryotic networks. Although the evolutionary mechanisms in early ancestors of prokaryotes and eukaryotes are quite similar, more novel and duplicated domain compositions with identical catalytic functions arise along the eukaryotic lineage. In contrast, prokaryotic enzymes become more versatile by catalyzing multiple reactions with similar chemical operations. Moreover, different metabolic pathways are enriched with distinct network evolution mechanisms. For instance, although the pathways of steroid biosynthesis, protein kinases, and glycosaminoglycan biosynthesis all constitute prominent features of animal-specific physiology, their evolution of domain architectures and catalytic functions follows distinct patterns. Steroid biosynthesis is enriched with reaction creations but retains a relatively conserved repertoire of domain compositions and proteins. Protein kinases retain conserved reactions but possess many novel domains and proteins. In contrast, glycosaminoglycan biosynthesis has high rates of reaction/protein creations and domain recruitments. Finally, we elicit and validate two general principles underlying the evolution of DPR networks: 1) duplicated enzyme proteins possess similar catalytic functions and 2) the majority of novel domains arise to catalyze novel reactions. These results shed new lights on the evolution of metabolic systems.

Segmentation of 3D microPET images of the rat brain via the hybrid gaussian mixture method with kernel density estimation

Segmentation of positron emission tomography (PET) is typically achieved using the K-Means method or other approaches. In preclinical and clinical applications, the K-Means method needs a prior estimation of parameters such as the number of clusters and appropriate initialized values. This work segments microPET images using a hybrid method combining the Gaussian mixture model (GMM) with kernel density estimation. Segmentation is crucial to registration of disordered 2-deoxy-2-fluoro-D-glucose (FDG) accumulation locations with functional diagnosis and to estimate standardized uptake values (SUVs) of region of interests (ROIs) in PET images. Therefore, simulation studies are conducted to apply spherical targets to evaluate segmentation accuracy based on Tanimoto's definition of similarity. The proposed method generates a higher degree of similarity than the K-Means method. The PET images of a rat brain are used to compare the segmented shape and area of the cerebral cortex by the K-Means method and the proposed method by volume rendering. The proposed method provides clearer and more detailed activity structures of an FDG accumulation location in the cerebral cortex than those by the K-Means method.

Constructing biological pathways by a two-step counting approach

Networks are widely used in biology to represent the relationships between genes and gene functions. In Boolean biological models, it is mainly assumed that there are two states to represent a gene: on-state and off-state. It is typically assumed that the relationship between two genes can be characterized by two kinds of pairwise relationships: similarity and prerequisite. Many approaches have been proposed in the literature to reconstruct biological relationships. In this article, we propose a two-step method to reconstruct the biological pathway when the binary array data have measurement error. For a pair of genes in a sample, the first step of this approach is to assign counting numbers for every relationship and select the relationship with counting number greater than a threshold. The second step is to calculate the asymptotic p-values for hypotheses of possible relationships and select relationships with a large p-value. This new method has the advantages of easy calculation for the counting numbers and simple closed forms for the p-value. The simulation study and real data example show that the two-step counting method can accurately reconstruct the biological pathway and outperform the existing methods. Compared with the other existing methods, this two-step method can provide a more accurate and efficient alternative approach for reconstructing the biological network.

Natural selection on cis and trans regulation in yeasts

Gene expression is regulated both by cis elements, which are DNA segments closely linked to the genes they regulate, and by trans factors, which are usually proteins capable of diffusing to unlinked genes. Understanding the patterns and sources of regulatory variation is crucial for understanding phenotypic and genome evolution. Here, we measure genome-wide allele-specific expression by deep sequencing to investigate the patterns of cis and trans expression variation between two strains of Saccharomyces cerevisiae. We propose a statistical modeling framework based on the binomial distribution that simultaneously addresses normalization of read counts derived from different parents and estimating the cis and trans expression variation parameters. We find that expression polymorphism in yeast is common for both cis and trans, though trans variation is more common. Constraint in expression evolution is correlated with other hallmarks of constraint, including gene essentiality, number of protein interaction partners, and constraint in amino acid substitution, indicating that both cis and trans polymorphism are clearly under purifying selection, though trans variation appears to be more sensitive to selective constraint. Comparing interspecific expression divergence between S. cerevisiae and S. paradoxus to our intraspecific variation suggests a significant departure from a neutral model of molecular evolution. A further examination of correlation between polymorphism and divergence within each category suggests that cis divergence is more frequently mediated by positive Darwinian selection than is trans divergence.

MicroPET reconstruction with random coincidence correction via a joint Poisson model

Positron emission tomography (PET) can provide in vivo, quantitative and functional information for diagnosis; however, PET image quality depends highly on a reconstruction algorithm. Iterative algorithms, such as the maximum likelihood expectation maximization (MLEM) algorithm, are rapidly becoming the standards for image reconstruction in emission-computed tomography. The conventional MLEM algorithm utilized the Poisson model in its system matrix, which is no longer valid for delay-subtraction of randomly corrected data. The aim of this study is to overcome this problem. The maximum likelihood estimation using the expectation maximum algorithm (MLE-EM) is adopted and modified to reconstruct microPET images using random correction from joint prompt and delay sinograms; this reconstruction method is called PDEM. The proposed joint Poisson model preserves Poisson properties without increasing the variance (noise) associated with random correction. The work here is an initial application/demonstration without applied normalization, scattering, attenuation, and arc correction. The coefficients of variation (CV) and full width at half-maximum (FWHM) values were utilized to compare the quality of reconstructed microPET images of physical phantoms acquired by filtered backprojection (FBP), ordered subsets-expected maximum (OSEM) and PDEM approaches. Experimental and simulated results demonstrate that the proposed PDEM produces better image quality than the FBP and OSEM approaches.

Multidimensional scaling for large genomic data sets

BACKGROUND

Multi-dimensional scaling (MDS) is aimed to represent high dimensional data in a low dimensional space with preservation of the similarities between data points. This reduction in dimensionality is crucial for analyzing and revealing the genuine structure hidden in the data. For noisy data, dimension reduction can effectively reduce the effect of noise on the embedded structure. For large data set, dimension reduction can effectively reduce information retrieval complexity. Thus, MDS techniques are used in many applications of data mining and gene network research. However, although there have been a number of studies that applied MDS techniques to genomics research, the number of analyzed data points was restricted by the high computational complexity of MDS. In general, a non-metric MDS method is faster than a metric MDS, but it does not preserve the true relationships. The computational complexity of most metric MDS methods is over O(N2), so that it is difficult to process a data set of a large number of genes N, such as in the case of whole genome microarray data.

RESULTS

We developed a new rapid metric MDS method with a low computational complexity, making metric MDS applicable for large data sets. Computer simulation showed that the new method of split-and-combine MDS (SC-MDS) is fast, accurate and efficient. Our empirical studies using microarray data on the yeast cell cycle showed that the performance of K-means in the reduced dimensional space is similar to or slightly better than that of K-means in the original space, but about three times faster to obtain the clustering results. Our clustering results using SC-MDS are more stable than those in the original space. Hence, the proposed SC-MDS is useful for analyzing whole genome data.

CONCLUSION

Our new method reduces the computational complexity from O(N3) to O(N) when the dimension of the feature space is far less than the number of genes N, and it successfully reconstructs the low dimensional representation as does the classical MDS. Its performance depends on the grouping method and the minimal number of the intersection points between groups. Feasible methods for grouping methods are suggested; each group must contain both neighboring and far apart data points. Our method can represent high dimensional large data set in a low dimensional space not only efficiently but also effectively.

Segmentation of cDNA microarray images by kernel density estimation

The segmentation of cDNA microarray spots is essential in analyzing the intensities of microarray images for biological and medical investigation. In this work, nonparametric methods using kernel density estimation are applied to segment two-channel cDNA microarray images. This approach groups pixels into both a foreground and a background. The segmentation performance of this model is tested and evaluated with reference to 16 microarray data. In particular, spike genes with various contents are spotted in a microarray to examine and evaluate the accuracy of the segmentation results. Duplicated design is implemented to evaluate the accuracy of the model. The results of this study demonstrate that this method can cluster pixels and estimate statistics regarding spots with high accuracy.

Method for identifying transcription factor binding sites in yeast

MOTIVATION

Identifying transcription factor binding sites (TFBSs) is helpful for understanding the mechanism of transcriptional regulation. The abundance and the diversity of genomic data provide an excellent opportunity for identifying TFBSs. Developing methods to integrate various types of data has become a major trend in this pursuit.

RESULTS

We develop a TFBS identification method, TFBSfinder, which utilizes several data sources, including DNA sequences, phylogenetic information, microarray data and ChIP-chip data. For a TF, TFBSfinder rigorously selects a set of reliable target genes and a set of non-target genes (as a background set) to find overrepresented and conserved motifs in target genes. A new metric for measuring the degree of conservation at a binding site across species and methods for clustering motifs and for inferring position weight matrices are proposed. For synthetic data and yeast cell cycle TFs, TFBSfinder identifies motifs that are highly similar to known consensuses. Moreover, TFBSfinder outperforms well-known methods.

AVAILABILITY

http://cg1.iis.sinica.edu.tw/~TFBSfinder/.

Statistical methods for identifying yeast cell cycle transcription factors

Knowing transcription factors (TFs) involved in the yeast cell cycle is helpful for understanding the regulation of yeast cell cycle genes. We therefore developed two methods for predicting (i) individual cell cycle TFs and (ii) synergistic TF pairs. The essential idea is that genes regulated by a cell cycle TF should have higher (lower, if it is a repressor) expression levels than genes not regulated by it during one or more phases of the cell cycle. This idea can also be used to identify synergistic interactions of TFs. Applying our methods to chromatin immunoprecipitation data and microarray data, we predict 50 cell cycle TFs and 80 synergistic TF pairs, including most known cell cycle TFs and synergistic TF pairs. Using these and published results, we describe the behaviors of 50 known or inferred cell cycle TFs in each cell cycle phase in terms of activation/repression and potential positive/negative interactions between TFs. In addition to the cell cycle, our methods are also applicable to other functions.

Explore biological pathways from noisy array data by directed acyclic Boolean networks

We consider the structure of directed acyclic Boolean (DAB) networks as a tool for exploring biological pathways. In a DAB network, the basic objects are binary elements and their Boolean duals. A DAB is characterized by two kinds of pairwise relations: similarity and prerequisite. The latter is a partial order relation, namely, the on-status of one element is necessary for the on-status of another element. A DAB network is uniquely determined by the state space of its elements. We arrange samples from the state space of a DAB network in a binary array and introduce a random mechanism of measurement error. Our inference strategy consists of two stages. First, we consider each pair of elements and try to identify their most likely relation. In the meantime, we assign a score, s-p-score, to this relation. Second, we rank the s-p-scores obtained from the first stage. We expect that relations with smaller s-p-scores are more likely to be true, and those with larger s-p-scores are more likely to be false. The key idea is the definition of s-scores (referring to similarity), p-scores (referring to prerequisite), and s-p-scores. As with classical statistical tests, control of false negatives and false positives are our primary concerns. We illustrate the method by a simulated example, the classical arginine biosynthetic pathway, and show some exploratory results on a published microarray expression dataset of yeast Saccharomyces cerevisiae obtained from experiments with activation and genetic perturbation of the pheromone response MAPK pathway.