A novel collaborative learning model for mixed dentition and fillings segmentation in panoramic radiographs.

INTRODUCTION: It is critical for dentists to identify and differentiate primary and permanent teeth, fillings, dental restorations and areas with pathological findings when reviewing dental radiographs to ensure that an accurate diagnosis is made and the optimal treatment can be planned. Unfortunately, dental radiographs are sometimes read incorrectly due to human error or low-quality images. While secondary or group review can help catch errors, many dentists work in practice alone and/or do not have time to review all of their patients' radiographs with another dentist. Artificial intelligence may facilitate the accurate interpretation of radiographs. To help support the review of panoramic radiographs, we developed a novel collaborative learning model that simultaneously identifies and differentiates primary and permanent teeth and detects fillings. METHODS: We used publicly accessible dental panoramic radiographic images and images obtained from the University of Missouri-Kansas City, School of Dentistry to develop and optimize two high-performance classifiers: (1) a system for tooth segmentation that can differentiate primary and permanent teeth and (2) a system to detect dental fillings. RESULTS: By utilizing these high-performance classifiers, we created models that can identify primary and permanent teeth (mean average precision [mAP] 95.32 % and performance [F-1] 92.50 %), as well as their associated dental fillings (mAP 91.53 % and F-1 91.00 %). We also designed a novel method for collaborative learning that utilizes these two classifiers to enhance recognition performance (mAP 94.09 % and F-1 93.41 %). CONCLUSIONS: Our model improves upon the existing machine learning models to simultaneously identify and differentiate primary and permanent teeth, and to identify any associated fillings. CLINICAL SIGNIFICANCE: Human error can lead to incorrect readings of panoramic radiographs. By developing artificial intelligence and machine learning methods to analyze panoramic radiographs, dentists can use this information to support their radiograph interpretations, help communicate the information to patients, and assist dental students learning to read radiographs.

Telehealth of cardiac devices for CVD treatment.

This review covers currently available cardiac implantable electronic devices (CIEDs) as well as updated progress in real-time monitoring techniques for CIEDs. A variety of implantable and wearable devices that can diagnose and monitor patients with cardiovascular diseases are summarized, and various working mechanisms and principles of monitoring techniques for Telehealth and mHealth are discussed. In addition, future research directions are presented based on the rapidly evolving research landscape including Artificial Intelligence (AI).

A case study of using natural language processing to extract consumer insights from tweets in American cities for public health crises.

BACKGROUND: The COVID-19 pandemic was a "wake up" call for public health agencies. Often, these agencies are ill-prepared to communicate with target audiences clearly and effectively for community-level activations and safety operations. The obstacle is a lack of data-driven approaches to obtaining insights from local community stakeholders. Thus, this study suggests a focus on listening at local levels given the abundance of geo-marked data and presents a methodological solution to extracting consumer insights from unstructured text data for health communication. METHODS: This study demonstrates how to combine human and Natural Language Processing (NLP) machine analyses to reliably extract meaningful consumer insights from tweets about COVID and the vaccine. This case study employed Latent Dirichlet Allocation (LDA) topic modeling, Bidirectional Encoder Representations from Transformers (BERT) emotion analysis, and human textual analysis and examined 180,128 tweets scraped by Twitter Application Programming Interface's (API) keyword function from January 2020 to June 2021. The samples came from four medium-sized American cities with larger populations of people of color. RESULTS: The NLP method discovered four topic trends: "COVID Vaccines," "Politics," "Mitigation Measures," and "Community/Local Issues," and emotion changes over time. The human textual analysis profiled the discussions in the selected four markets to add some depth to our understanding of the uniqueness of the different challenges experienced. CONCLUSIONS: This study ultimately demonstrates that our method used here could efficiently reduce a large amount of community feedback (e.g., tweets, social media data) by NLP and ensure contextualization and richness with human interpretation. Recommendations on communicating vaccination are offered based on the findings: (1) the strategic objective should be empowering the public; (2) the message should have local relevance; and, (3) communication needs to be timely.

ETSNet: A deep neural network for EEG-based temporal-spatial pattern recognition in psychiatric disorder and emotional distress classification.

The use of EEG for evaluating and diagnosing neurological abnormalities related to psychiatric diseases and identifying human emotions has been improved by deep learning advancements. This research aims to categorize individuals with schizophrenia (SZ), their biological relatives (REL), and healthy controls (HC) using resting EEG brain source signal data defined by regions of interest (ROIs). The proposed solution is a deep neural network for the cortical source signals of the ROIs, incorporating a Squeeze-and-Excitation Block and multiple CNNs designed for eyes-open and closed resting states. The model, called EEG Temporal Spatial Network (ETSNet), has two variants: ETSNets and ETSNetf. Two evaluations were conducted to show the effectiveness of the proposed model. The average accuracy for the classification of SZ, REL, and HC using EEG resting data was 99.57% (ETSNetf), and the average accuracy for the classification of eyes-open (EO) and eyes-closed (EC) resting states was 93.15% (ETSNets). An ablation study was also conducted using two public datasets for intellectual and developmental disorders and emotional states, showing improved classification accuracy compared to advanced EEG classification algorithms when using ETSNets.

Multimodal ensemble model for Alzheimer's disease conversion prediction from Early Mild Cognitive Impairment subjects.

Alzheimer's Disease (AD) is the most common type of dementia. Predicting the conversion to Alzheimer's from the mild cognitive impairment (MCI) stage is a complex problem that has been studied extensively. This study centers on individualized EMCI (the earliest MCI subset) to AD conversion prediction on multimodal data such as diffusion tensor imaging (DTI) scans and electronic health records (EHR) for their patients using the combination of both a balanced random forest model alongside a convolutional neural network (CNN) model. Our random forest model leverages EHR's patient biometric and neuropsychiatric test score features, while our CNN model uses the patient's diffusion tensor imaging (DTI) scans for conversion prediction. To accomplish this, 383 Early Mild Cognitive Impairment (EMCI) patients were collected from the Alzheimer's Disease Neuroimaging Initiative (ADNI). Within this set, 49 patients would eventually convert to AD (EMCI_C), whereas the remaining 335 did not convert (EMCI_NC). For the EHR-based classifier, 288 patients were used to train the random forest model, with 95 set aside for testing. For the CNN classifier, 405 DTI images were collected across 90 distinct patients. Nine clinical features were selected to be combined with the visual predictor. Due to the imbalanced classes, oversampling was performed for the clinical features and augmentation for the DTI images. A grid search algorithm is also used to determine the ideal weighting between our two models. Our results indicate that an ensemble model was effective (98.81% accuracy) at EMCI to AD conversion prediction. Additionally, our ensemble model provides explainability as feature importance can be assessed at both the model and individual prediction levels. Therefore, this ensemble model could serve as a diagnostic support tool or a means for identifying clinical trial candidates.

Stacked Ensemble Network to Assess the Structural Variations in Retina: A Bio-marker for Early Disease Diagnosis.

The retina is a unique tissue that extends the human brain in transmitting the incoming light into neural spikes. Researchers collaborating with domain experts proposed numerous deep networks to extract vessels from the retina; however, these techniques have the least response for micro-vessels. The proposed method has developed a stacked ensemble network approach with deep neural architectures for precise vessel extraction. Our method has used bi-directional LSTM for filling gaps in dis-joint vessels and applied W-Net for boundary refinement and emphasizing local regions to achieve better results for micro-vessels extraction. The platform has combined the strength of various networks to improve the automated screening process and has shown promising results on benchmark datasets.

Incremental Learning for Panoramic Radiograph Segmentation.

This study aimed to determine a fundamental method for the automated detection and treatment of dental and orthodontic problems. Manual intervention is inefficient and prone to human error in detecting anomalies. Deep learning was used to identify a solution to this problem. We proposed leveraging incremental learning approaches using Mask RCNN as backbone networks on small datasets to construct a more accurate model from automatically labeled data. The knowledge acquired at one stage of education is carried over to the subsequent stage. By incorporating newly annotated data, transfer learning improved the model's performance. Despite the data scarcity issues inherent in radiograph image collection, the findings for filling and tooth segmentation tasks were encouraging and adequate. We compared our results to prior research to optimize the performance of our proposed method.

Collaborative deep learning model for tooth segmentation and identification using panoramic radiographs.

Panoramic radiographs are an integral part of effective dental treatment planning, supporting dentists in identifying impacted teeth, infections, malignancies, and other dental issues. However, screening for anomalies solely based on a dentist's assessment may result in diagnostic inconsistency, posing difficulties in developing a successful treatment plan. Recent advancements in deep learning-based segmentation and object detection algorithms have enabled the provision of predictable and practical identification to assist in the evaluation of a patient's mineralized oral health, enabling dentists to construct a more successful treatment plan. However, there has been a lack of efforts to develop collaborative models that enhance learning performance by leveraging individual models. The article describes a novel technique for enabling collaborative learning by incorporating tooth segmentation and identification models created independently from panoramic radiographs. This collaborative technique permits the aggregation of tooth segmentation and identification to produce enhanced results by recognizing and numbering existing teeth (up to 32 teeth). The experimental findings indicate that the proposed collaborative model is significantly more effective than individual learning models (e.g., 98.77% vs. 96% and 98.44% vs.91% for tooth segmentation and recognition, respectively). Additionally, our models outperform the state-of-the-art segmentation and identification research. We demonstrated the effectiveness of collaborative learning in detecting and segmenting teeth in a variety of complex situations, including healthy dentition, missing teeth, orthodontic treatment in progress, and dentition with dental implants.

Feature-Based Fusion Using CNN for Lung and Heart Sound Classification.

Lung or heart sound classification is challenging due to the complex nature of audio data, its dynamic properties of time, and frequency domains. It is also very difficult to detect lung or heart conditions with small amounts of data or unbalanced and high noise in data. Furthermore, the quality of data is a considerable pitfall for improving the performance of deep learning. In this paper, we propose a novel feature-based fusion network called FDC-FS for classifying heart and lung sounds. The FDC-FS framework aims to effectively transfer learning from three different deep neural network models built from audio datasets. The innovation of the proposed transfer learning relies on the transformation from audio data to image vectors and from three specific models to one fused model that would be more suitable for deep learning. We used two publicly available datasets for this study, i.e., lung sound data from ICHBI 2017 challenge and heart challenge data. We applied data augmentation techniques, such as noise distortion, pitch shift, and time stretching, dealing with some data issues in these datasets. Importantly, we extracted three unique features from the audio samples, i.e., Spectrogram, MFCC, and Chromagram. Finally, we built a fusion of three optimal convolutional neural network models by feeding the image feature vectors transformed from audio features. We confirmed the superiority of the proposed fusion model compared to the state-of-the-art works. The highest accuracy we achieved with FDC-FS is 99.1% with Spectrogram-based lung sound classification while 97% for Spectrogram and Chromagram based heart sound classification.

AM-UNet: automated mini 3D end-to-end U-net based network for brain claustrum segmentation.

Recent advances in deep learning (DL) have provided promising solutions to medical image segmentation. Among existing segmentation approaches, the U-Net-based methods have been used widely. However, very few U-Net-based studies have been conducted on automatic segmentation of the human brain claustrum (CL). The CL segmentation is challenging due to its thin, sheet-like structure, heterogeneity of its image modalities and formats, imperfect labels, and data imbalance. We propose an automatic optimized U-Net-based 3D segmentation model, called AM-UNet, designed as an end-to-end process of the pre and post-process techniques and a U-Net model for CL segmentation. It is a lightweight and scalable solution which has achieved the state-of-the-art accuracy for automatic CL segmentation on 3D magnetic resonance images (MRI). On the T1/T2 combined MRI CL dataset, AM-UNet has obtained excellent results, including Dice, Intersection over Union (IoU), and Intraclass Correlation Coefficient (ICC) scores of 82%, 70%, and 90%, respectively. We have conducted the comparative evaluation of AM-UNet with other pre-existing models for segmentation on the MRI CL dataset. As a result, medical experts confirmed the superiority of the proposed AM-UNet model for automatic CL segmentation. The source code and model of the AM-UNet project is publicly available on GitHub: https://github.com/AhmedAlbishri/AM-UNET.

Event-Driven Deep Learning for Edge Intelligence (EDL-EI).

Edge intelligence (EI) has received a lot of interest because it can reduce latency, increase efficiency, and preserve privacy. More significantly, as the Internet of Things (IoT) has proliferated, billions of portable and embedded devices have been interconnected, producing zillions of gigabytes on edge networks. Thus, there is an immediate need to push AI (artificial intelligence) breakthroughs within edge networks to achieve the full promise of edge data analytics. EI solutions have supported digital technology workloads and applications from the infrastructure level to edge networks; however, there are still many challenges with the heterogeneity of computational capabilities and the spread of information sources. We propose a novel event-driven deep-learning framework, called EDL-EI (event-driven deep learning for edge intelligence), via the design of a novel event model by defining events using correlation analysis with multiple sensors in real-world settings and incorporating multi-sensor fusion techniques, a transformation method for sensor streams into images, and lightweight 2-dimensional convolutional neural network (CNN) models. To demonstrate the feasibility of the EDL-EI framework, we presented an IoT-based prototype system that we developed with multiple sensors and edge devices. To verify the proposed framework, we have a case study of air-quality scenarios based on the benchmark data provided by the USA Environmental Protection Agency for the most polluted cities in South Korea and China. We have obtained outstanding predictive accuracy (97.65% and 97.19%) from two deep-learning models on the cities' air-quality patterns. Furthermore, the air-quality changes from 2019 to 2020 have been analyzed to check the effects of the COVID-19 pandemic lockdown.

Parametric Optimization of 3D Printed Hydrogel-Based Cardiovascular Stent.

PURPOSE: This study aimed to develop personalized biodegradable stent (BDS) for the treatment of coronary heart disease. Three-dimensional (3D) printing technique has offered easy and fast fabrication of BDS with enhanced reproducibility and efficacy. METHODS: A variety of BDS were printed with 3 types of hydrogel (~5 ml) resources (10%w/v sodium alginate (SA), 10%w/v cysteine-sodium alginate (SA-CYS), and 10%w/v cysteine-sodium alginate with 0.4%w/v PLA-nanofibers (SA-CYS-NF)) dispersed from an 22G print head nozzle attached to the BD-syringe. The printability of hydrogels into 3D structures was examined based on such variables as hydrogel's viscosity, printing distance, printing speed and the nozzle size. RESULTS: It was demonstrated that alginate composition (10%w/v) offered BDS with sufficient viscosity that defined the thickness and swelling ratio of the stent struts. The thickness of the strut was found to be 338.7 ± 29.3 µm, 262.5 ± 14.7 µm and 237.1 ± 14.7 µm for stents made of SA, SA-CYS and SA-CYS-NF, respectively. SA-CYS-NF stent displayed the highest swelling ratio of 38.8 ± 2.9% at the initial 30 min, whereas stents made of SA and SA-CYS had 23.1 ± 2.4% and 22.0 ± 2.4%, respectively. CONCLUSION: The printed stents had sufficient mechanical strength and were stable against pseudo-physiological wall shear stress. An addition of nanofibers to alginate hydrogel significantly enhanced the biodegradation rates of the stents. In vitro cell culture studies revealed that stents had no cytotoxic effects on human umbilical vein endothelial cells (HUVECs) and Raw 264.7 cells (i.e., Monocyte/macrophage-like cells), supporting that stents are biocompatible and can be explored for future clinical applications.

Random forest model for feature-based Alzheimer's disease conversion prediction from early mild cognitive impairment subjects.

Alzheimer's Disease (AD) conversion prediction from the mild cognitive impairment (MCI) stage has been a difficult challenge. This study focuses on providing an individualized MCI to AD conversion prediction using a balanced random forest model that leverages clinical data. In order to do this, 383 Early Mild Cognitive Impairment (EMCI) patients were gathered from the Alzheimer's Disease Neuroimaging Initiative (ADNI). Of these patients, 49 would eventually convert to AD (EMCI_C), whereas the remaining 334 did not convert (EMCI_NC). All of these patients were split randomly into training and testing data sets with 95 patients reserved for testing. Nine clinical features were selected, comprised of a mix of demographic, brain volume, and cognitive testing variables. Oversampling was then performed in order to balance the initially imbalanced classes prior to training the model with 1000 estimators. Our results showed that a random forest model was effective (93.6% accuracy) at predicting the conversion of EMCI patients to AD based on these clinical features. Additionally, we focus on explainability by assessing the importance of each clinical feature. Our model could impact the clinical environment as a tool to predict the conversion to AD from a prodromal stage or to identify ideal candidates for clinical trials.

A Data-Mining Approach for the Quantitative Assessment of Physicochemical Properties of Molecular Compounds in the Skin Flux.

This paper aimed to provide an insight into the mechanism of transdermal penetration of drug molecules with respect to their physicochemical properties, such as solubility (S), the presence of enantiomer (ET) and logarithm of octanol-water partition coefficient (log P), molecular weight (MW), and melting point (MP). Propionic acid derivatives were evaluated for their flux through full-thickness skin excised from hairless mice upon being delivered from silicone-based pressure-sensitive adhesive (PSA) matrices in the presence or absence of various enhancers. The skin fluxes of model compounds were calculated based on the data obtained using the method engaged with the diffusion cell system. The statistical design of experiments (DoE) based on the factorial approach was used to find variables that have a significant impact on the outcomes. For the prediction of skin flux, a quantitative equation was derived using the data-mining approach on the relationship between skin permeation of model compounds (~125 mg/ml) and involved physicochemical variables. The most influential variables for the skin flux of propionic acid derivatives were the melting point (0.97) followed by the presence of enantiomer (0.95), molecular mass (0.93), log P values (0.86), and aqueous solubility (0.80). It was concluded that the skin flux of molecular compounds can be predicted based on the relationship between their physicochemical properties and the interaction with cofactors including additives and enhancers in the vehicles.

Protein structure search to support the development of protein structure prediction methods.

Protein structure prediction is a long-standing unsolved problem in molecular biology that has seen renewed interest with the recent success of deep learning with AlphaFold at CASP13. While developing and evaluating protein structure prediction methods, researchers may want to identify the most similar known structures to their predicted structures. These predicted structures often have low sequence and structure similarity to known structures. We show how RUPEE, a purely geometric protein structure search, is able to identify the structures most similar to structure predictions, regardless of how they vary from known structures, something existing protein structure searches struggle with. RUPEE accomplishes this through the use of a novel linear encoding of protein structures as a sequence of residue descriptors. Using a fast Needleman-Wunsch algorithm, RUPEE is able to perform alignments on the sequences of residue descriptors for every available structure. This is followed by a series of increasingly accurate structure alignments from TM-align alignments initialized with the Needleman-Wunsch residue descriptor alignments to standard TM-align alignments of the final results. By using alignment normalization effectively at each stage, RUPEE also can execute containment searches in addition to full-length searches to identify structural motifs within proteins. We compare the results of RUPEE to the protein structure searches mTM-align, SSM, CATHEDRAL, and VAST using a benchmark derived from the protein structure predictions submitted to CASP13. RUPEE identifies better alignments on average with respect to TM-score as well as scores specific to SSM and CATHEDRAL, Q-score and SSAP-score, respectively.

Deep Learning in Personalization of Cardiovascular Stents.

Deep learning (DL) application has demonstrated its enormous potential in accomplishing biomedical tasks, such as vessel segmentation, brain visualization, and speech recognition. This review article has mainly covered recent advances in the principles of DL algorithms, existing DL software, and designing strategies of DL models. Latest progresses in cardiovascular devices, especially DL-based cardiovascular stent used for angioplasty, differential and advanced diagnostic means, and the treatment outcomes involved with coronary artery disease (CAD), are discussed. Also presented is DL-based discovery of new materials and future medical technologies that will facilitate the development of tailored and personalized treatment strategies by identifying and forecasting individual impending risks of cardiovascular diseases.

Early temporal characteristics of elderly patient cognitive impairment in electronic health records.

BACKGROUND: The aging population has led to an increase in cognitive impairment (CI) resulting in significant costs to patients, their families, and society. A research endeavor on a large cohort to better understand the frequency and severity of CI is urgent to respond to the health needs of this population. However, little is known about temporal trends of patient health functions (i.e., activity of daily living [ADL]) and how these trends are associated with the onset of CI in elderly patients. Also, the use of a rich source of clinical free text in electronic health records (EHRs) to facilitate CI research has not been well explored. The aim of this study is to characterize and better understand early signals of elderly patient CI by examining temporal trends of patient ADL and analyzing topics of patient medical conditions in clinical free text using topic models. METHODS: The study cohort consists of physician-diagnosed CI patients (n = 1,435) and cognitively unimpaired (CU) patients (n = 1,435) matched by age and sex, selected from patients 65 years of age or older at the time of enrollment in the Mayo Clinic Biobank. A corpus analysis was performed to examine the basic statistics of event types and practice settings where the physician first diagnosed CI. We analyzed the distribution of ADL in three different age groups over time before the development of CI. Furthermore, we applied three different topic modeling approaches on clinical free text to examine how patients' medical conditions change over time when they were close to CI diagnosis. RESULTS: The trajectories of ADL deterioration became steeper in CI patients than CU patients approximately 1 to 1.5 year(s) before the actual physician diagnosis of CI. The topic modeling showed that the topic terms were mostly correlated and captured the underlying semantics relevant to CI when approaching to CI diagnosis. CONCLUSIONS: There exist notable differences in temporal trends of basic and instrumental ADL between CI and CU patients. The trajectories of certain individual ADL, such as bathing and responsibility of own medication, were closely associated with CI development. The topic terms obtained by topic modeling methods from clinical free text have a potential to show how CI patients' conditions evolve and reveal overlooked conditions when they close to CI diagnosis.

RUPEE: A fast and accurate purely geometric protein structure search.

Given the close relationship between protein structure and function, protein structure searches have long played an established role in bioinformatics. Despite their maturity, existing protein structure searches either use simplifying assumptions or compromise between fast response times and quality of results. These limitations can prevent the easy and efficient exploration of relationships between protein structures, which is the norm in other areas of inquiry. To address these limitations we have developed RUPEE, a fast and accurate purely geometric structure search combining techniques from information retrieval and big data with a novel approach to encoding sequences of torsion angles. Comparing our results to the output of mTM, SSM, and the CATHEDRAL structural scan, it is clear that RUPEE has set a new bar for purely geometric big data approaches to protein structure searches. RUPEE in top-aligned mode produces equal or better results than the best available protein structure searches, and RUPEE in fast mode demonstrates the fastest response times coupled with high quality results. The RUPEE protein structure search is available at https://ayoubresearch.com. Code and data are available at https://github.com/rayoub/rupee.

Contextual Word Embeddings and Topic Modeling in Healthy Dieting and Obesity.

An alarming proportion of the US population is overweight. Obesity increases the risk of illnesses such as diabetes and cardiovascular diseases. In this paper, we propose the Contextual Word Embeddings (ContWEB) framework that aims to build contextual word embeddings on the relationship between obesity and healthy eating from the crowd domain (Twitter) and the expert domain (PubMed). For this purpose, our work is based on a pipeline model that consists of a chain of processing elements as follows: (1) to use term frequency and inverse document frequency (TF-IDF) and Word2Vec in the data collected from the crowd and expert domains; (2) to apply natural language processing (NLP) algorithms to the corpus; (3) to construct social word embeddings by sentiment analysis; (4) to discover the contextual word embeddings using co-occurrence and conditional probability; (5) to find an optimal number of topics in a topic modeling with the obesity and healthy dieting corpus; (6) to extract latent features extracted using Latent Dirichlet Allocation (LDA). The ContWEB framework has been implemented on Apache Spark and TensorFlow platforms. We have evaluated the ContWEB framework in terms of the effectiveness in contextual word embeddings constructed from the crowd and the expert domains. We conclude that the ContWEB framework would be useful in enhancing the decision-making process for healthy eating and obesity prevention.

Augmented reality for personalized nanomedicines.

As our understanding of onset and progress of diseases at the genetic and molecular level rapidly progresses, the potential of advanced technologies, such as 3D-printing, Socially-Assistive Robots (SARs) or augmented reality (AR), that are applied to personalized nanomedicines (PNMs) to alleviate pathological conditions, has become more prominent. Among advanced technologies, AR in particular has the greatest potential to address those challenges and facilitate the translation of PNMs into formidable clinical application of personalized therapy. As AR is about to adapt additional new methods, such as speech, voice recognition, eye tracing and motion tracking, to enable interaction with host response or biological systems in 3-D space, a combination of multiple approaches to accommodate varying environmental conditions, such as public noise and atmosphere brightness, will be explored to improve its therapeutic outcomes in clinical applications. For instance, AR glasses still being developed by Facebook or Microsoft will serve as new platform that can provide people with the health information they are interested in or various measures through which they can interact with medical services. This review has addressed the current progress and impact of AR on PNMs and its application to the biomedical field. Special emphasis is placed on the application of AR based PNMs to the treatment strategies against senior care, drug addiction and medication adherence.