HARNet in deep learning approach-a systematic survey.

A comprehensive examination of human action recognition (HAR) methodologies situated at the convergence of deep learning and computer vision is the subject of this article. We examine the progression from handcrafted feature-based approaches to end-to-end learning, with a particular focus on the significance of large-scale datasets. By classifying research paradigms, such as temporal modelling and spatial features, our proposed taxonomy illuminates the merits and drawbacks of each. We specifically present HARNet, an architecture for Multi-Model Deep Learning that integrates recurrent and convolutional neural networks while utilizing attention mechanisms to improve accuracy and robustness. The VideoMAE v2 method ( https://github.com/OpenGVLab/VideoMAEv2 ) has been utilized as a case study to illustrate practical implementations and obstacles. For researchers and practitioners interested in gaining a comprehensive understanding of the most recent advancements in HAR as they relate to computer vision and deep learning, this survey is an invaluable resource.

A decision support system based on recurrent neural networks to predict medication dosage for patients with Parkinson's disease.

Using deep learning has demonstrated significant potential in making informed decisions based on clinical evidence. In this study, we deal with optimizing medication and quantitatively present the role of deep learning in predicting the medication dosage for patients with Parkinson's disease (PD). The proposed method is based on recurrent neural networks (RNNs) and tries to predict the dosage of five critical medication types for PD, including levodopa, dopamine agonists, monoamine oxidase-B inhibitors, catechol-O-methyltransferase inhibitors, and amantadine. Recurrent neural networks have memory blocks that retain crucial information from previous patient visits. This feature is helpful for patients with PD, as the neurologist can refer to the patient's previous state and the prescribed medication to make informed decisions. We employed data from the Parkinson's Progression Markers Initiative. The dataset included information on the Unified Parkinson's Disease Rating Scale, Activities of Daily Living, Hoehn and Yahr scale, demographic details, and medication use logs for each patient. We evaluated several models, such as multi-layer perceptron (MLP), Simple-RNN, long short-term memory (LSTM), and gated recurrent units (GRU). Our analysis found that recurrent neural networks (LSTM and GRU) performed the best. More specifically, when using LSTM, we were able to predict levodopa and dopamine agonist dosage with a mean squared error of 0.009 and 0.003, mean absolute error of 0.062 and 0.030, root mean square error of 0.099 and 0.053, and R-squared of 0.514 and 0.711, respectively.

High sensitivity methods for automated rib fracture detection in pediatric radiographs.

Rib fractures are highly predictive of non-accidental trauma in children under 3 years old. Rib fracture detection in pediatric radiographs is challenging because fractures can be obliquely oriented to the imaging detector, obfuscated by other structures, incomplete, and non-displaced. Prior studies have shown up to two-thirds of rib fractures may be missed during initial interpretation. In this paper, we implemented methods for improving the sensitivity (i.e. recall) performance for detecting and localizing rib fractures in pediatric chest radiographs to help augment performance of radiology interpretation. These methods adapted two convolutional neural network (CNN) architectures, RetinaNet and YOLOv5, and our previously proposed decision scheme, "avalanche decision", that dynamically reduces the acceptance threshold for proposed regions in each image. Additionally, we present contributions of using multiple image pre-processing and model ensembling techniques. Using a custom dataset of 1109 pediatric chest radiographs manually labeled by seven pediatric radiologists, we performed 10-fold cross-validation and reported detection performance using several metrics, including F2 score which summarizes precision and recall for high-sensitivity tasks. Our best performing model used three ensembled YOLOv5 models with varied input processing and an avalanche decision scheme, achieving an F2 score of 0.725 ± 0.012. Expert inter-reader performance yielded an F2 score of 0.732. Results demonstrate that our combination of sensitivity-driving methods provides object detector performance approaching the capabilities of expert human readers, suggesting that these methods may provide a viable approach to identify all rib fractures.

Remote sensing image information extraction based on Compensated Fuzzy Neural Network and big data analytics.

Medical imaging AI systems and big data analytics have attracted much attention from researchers of industry and academia. The application of medical imaging AI systems and big data analytics play an important role in the technology of content based remote sensing (CBRS) development. Environmental data, information, and analysis have been produced promptly using remote sensing (RS). The method for creating a useful digital map from an image data set is called image information extraction. Image information extraction depends on target recognition (shape and color). For low-level image attributes like texture, Classifier-based Retrieval(CR) techniques are ineffective since they categorize the input images and only return images from the determined classes of RS. The issues mentioned earlier cannot be handled by the existing expertise based on a keyword/metadata remote sensing data service model. To get over these restrictions, Fuzzy Class Membership-based Image Extraction (FCMIE), a technology developed for Content-Based Remote Sensing (CBRS), is suggested. The compensation fuzzy neural network (CFNN) is used to calculate the category label and fuzzy category membership of the query image. Use a basic and balanced weighted distance metric. Feature information extraction (FIE) enhances remote sensing image processing and autonomous information retrieval of visual content based on time-frequency meaning, such as color, texture and shape attributes of images. Hierarchical nested structure and cyclic similarity measure produce faster queries when searching. The experiment's findings indicate that applying the proposed model can have favorable outcomes for assessment measures, including Ratio of Coverage, average means precision, recall, and efficiency retrieval that are attained more effectively than the existing CR model. In the areas of feature tracking, climate forecasting, background noise reduction, and simulating nonlinear functional behaviors, CFNN has a wide range of RS applications. The proposed method CFNN-FCMIE achieves a minimum range of 4-5% for all three feature vectors, sample mean and comparison precision-recall ratio, which gives better results than the existing classifier-based retrieval model. This work provides an important reference for medical imaging artificial intelligence system and big data analysis.

Biodegradation of ciprofloxacin using machine learning tools: Kinetics and modelling.

Recently, the rampant administration of antibiotics and their synthetic organic constitutes have exacerbated adverse effects on ecosystems, affecting the health of animals, plants, and humans by promoting the emergence of extreme multidrug-resistant bacteria (XDR), antibiotic resistance bacterial variants (ARB), and genes (ARGs). The constraints, such as high costs, by-product formation, etc., associated with the physico-chemical treatment process limit their efficacy in achieving efficient wastewater remediation. Biodegradation is a cost-effective, energy-saving, sustainable alternative for removing emerging organic pollutants from environmental matrices. In view of the same, the current study aims to explore the biodegradation of ciprofloxacin using microbial consortia via metabolic pathways. The optimal parameters for biodegradation were assessed by employing machine learning tools, viz. Artificial Neural Network (ANN) and statistical optimization tool (Response Surface Methodology, RSM) using the Box-Behnken design (BBD). Under optimal culture conditions, the designed bacterial consortia degraded ciprofloxacin with 95.5% efficiency, aligning with model prediction results, i.e., 95.20% (RSM) and 94.53% (ANN), respectively. Thus, befitting amendments to the biodegradation process can augment efficiency and lead to a greener solution for antibiotic degradation from aqueous media.

DeepReg: a deep learning hybrid model for predicting transcription factors in eukaryotic and prokaryotic genomes.

Deep learning models (DLMs) have gained importance in predicting, detecting, translating, and classifying a diversity of inputs. In bioinformatics, DLMs have been used to predict protein structures, transcription factor-binding sites, and promoters. In this work, we propose a hybrid model to identify transcription factors (TFs) among prokaryotic and eukaryotic protein sequences, named Deep Regulation (DeepReg) model. Two architectures were used in the DL model: a convolutional neural network (CNN), and a bidirectional long-short-term memory (BiLSTM). DeepReg reached a precision of 0.99, a recall of 0.97, and an F1-score of 0.98. The quality of our predictions, the bias-variance trade-off approach, and the characterization of new TF predictions were evaluated and compared against those produced by DeepTFactor, as well as against experimental data from three model organisms. Predictions based on our DLM tended to exhibit less variance and bias than those from DeepTFactor, thus increasing reliability and decreasing overfitting.

Machine learning and new insights for breast cancer diagnosis.

Breast cancer (BC) is the most prominent form of cancer among females all over the world. The current methods of BC detection include X-ray mammography, ultrasound, computed tomography, magnetic resonance imaging, positron emission tomography and breast thermographic techniques. More recently, machine learning (ML) tools have been increasingly employed in diagnostic medicine for its high efficiency in detection and intervention. The subsequent imaging features and mathematical analyses can then be used to generate ML models, which stratify, differentiate and detect benign and malignant breast lesions. Given its marked advantages, radiomics is a frequently used tool in recent research and clinics. Artificial neural networks and deep learning (DL) are novel forms of ML that evaluate data using computer simulation of the human brain. DL directly processes unstructured information, such as images, sounds and language, and performs precise clinical image stratification, medical record analyses and tumour diagnosis. Herein, this review thoroughly summarizes prior investigations on the application of medical images for the detection and intervention of BC using radiomics, namely DL and ML. The aim was to provide guidance to scientists regarding the use of artificial intelligence and ML in research and the clinic.

Automatic brain-tumor diagnosis using cascaded deep convolutional neural networks with symmetric U-Net and asymmetric residual-blocks.

The use of various kinds of magnetic resonance imaging (MRI) techniques for examining brain tissue has increased significantly in recent years, and manual investigation of each of the resulting images can be a time-consuming task. This paper presents an automatic brain-tumor diagnosis system that uses a CNN for detection, classification, and segmentation of glioblastomas; the latter stage seeks to segment tumors inside glioma MRI images. The structure of the developed multi-unit system consists of two stages. The first stage is responsible for tumor detection and classification by categorizing brain MRI images into normal, high-grade glioma (glioblastoma), and low-grade glioma. The uniqueness of the proposed network lies in its use of different levels of features, including local and global paths. The second stage is responsible for tumor segmentation, and skip connections and residual units are used during this step. Using 1800 images extracted from the BraTS 2017 dataset, the detection and classification stage was found to achieve a maximum accuracy of 99%. The segmentation stage was then evaluated using the Dice score, specificity, and sensitivity. The results showed that the suggested deep-learning-based system ranks highest among a variety of different strategies reported in the literature.

Improving speech depression detection using transfer learning with wav2vec 2.0 in low-resource environments.

Depression, a pervasive global mental disorder, profoundly impacts daily lives. Despite numerous deep learning studies focused on depression detection through speech analysis, the shortage of annotated bulk samples hampers the development of effective models. In response to this challenge, our research introduces a transfer learning approach for detecting depression in speech, aiming to overcome constraints imposed by limited resources. In the context of feature representation, we obtain depression-related features by fine-tuning wav2vec 2.0. By integrating 1D-CNN and attention pooling structures, we generate advanced features at the segment level, thereby enhancing the model's capability to capture temporal relationships within audio frames. In the realm of prediction results, we integrate LSTM and self-attention mechanisms. This incorporation assigns greater weights to segments associated with depression, thereby augmenting the model's discernment of depression-related information. The experimental results indicate that our model has achieved impressive F1 scores, reaching 79% on the DAIC-WOZ dataset and 90.53% on the CMDC dataset. It outperforms recent baseline models in the field of speech-based depression detection. This provides a promising solution for effective depression detection in low-resource environments.

Diagnosis model of early Pneumocystis jirovecii pneumonia based on convolutional neural network: a comparison with traditional PCR diagnostic method.

BACKGROUND: Pneumocystis jirovecii pneumonia (PJP) is an interstitial pneumonia caused by pneumocystis jirovecii (PJ). The diagnosis of PJP primarily relies on the detection of the pathogen from lower respiratory tract specimens. However, it faces challenges such as difficulty in obtaining specimens and low detection rates. In the clinical diagnosis process, it is necessary to combine clinical symptoms, serological test results, chest Computed tomography (CT) images, molecular biology techniques, and metagenomics next-generation sequencing (mNGS) for comprehensive analysis. PURPOSE: This study aims to overcome the limitations of traditional PJP diagnosis methods and develop a non-invasive, efficient, and accurate diagnostic approach for PJP. By using this method, patients can receive early diagnosis and treatment, effectively improving their prognosis. METHODS: We constructed an intelligent diagnostic model for PJP based on the different Convolutional Neural Networks. Firstly, we used the Convolutional Neural Network to extract CT image features from patients. Then, we fused the CT image features with clinical information features using a feature fusion function. Finally, the fused features were input into the classification network to obtain the patient's diagnosis result. RESULTS: In this study, for the diagnosis of PJP, the accuracy of the traditional PCR diagnostic method is 77.58%, while the mean accuracy of the optimal diagnostic model based on convolutional neural networks is 88.90%. CONCLUSION: The accuracy of the diagnostic method proposed in this paper is 11.32% higher than that of the traditional PCR diagnostic method. The method proposed in this paper is an efficient, accurate, and non-invasive early diagnosis approach for PJP.

Classification and counting of cells in brightfield microscopy images: an application of convolutional neural networks.

Microscopy is integral to medical research, facilitating the exploration of various biological questions, notably cell quantification. However, this process's time-consuming and error-prone nature, attributed to human intervention or automated methods usually applied to fluorescent images, presents challenges. In response, machine learning algorithms have been integrated into microscopy, automating tasks and constructing predictive models from vast datasets. These models adeptly learn representations for object detection, image segmentation, and target classification. An advantageous strategy involves utilizing unstained images, preserving cell integrity and enabling morphology-based classification-something hindered when fluorescent markers are used. The aim is to introduce a model proficient in classifying distinct cell lineages in digital contrast microscopy images. Additionally, the goal is to create a predictive model identifying lineage and determining optimal quantification of cell numbers. Employing a CNN machine learning algorithm, a classification model predicting cellular lineage achieved a remarkable accuracy of 93%, with ROC curve results nearing 1.0, showcasing robust performance. However, some lineages, namely SH-SY5Y (78%), HUH7_mayv (85%), and A549 (88%), exhibited slightly lower accuracies. These outcomes not only underscore the model's quality but also emphasize CNNs' potential in addressing the inherent complexities of microscopic images.

TEC-miTarget: enhancing microRNA target prediction based on deep learning of ribonucleic acid sequences.

BACKGROUND: MicroRNAs play a critical role in regulating gene expression by binding to specific target sites within gene transcripts, making the identification of microRNA targets a prominent focus of research. Conventional experimental methods for identifying microRNA targets are both time-consuming and expensive, prompting the development of computational tools for target prediction. However, the existing computational tools exhibit limited performance in meeting the demands of practical applications, highlighting the need to improve the performance of microRNA target prediction models. RESULTS: In this paper, we utilize the most popular natural language processing and computer vision technologies to propose a novel approach, called TEC-miTarget, for microRNA target prediction based on transformer encoder and convolutional neural networks. TEC-miTarget treats RNA sequences as a natural language and encodes them using a transformer encoder, a widely used encoder in natural language processing. It then combines the representations of a pair of microRNA and its candidate target site sequences into a contact map, which is a three-dimensional array similar to a multi-channel image. Therefore, the contact map's features are extracted using a four-layer convolutional neural network, enabling the prediction of interactions between microRNA and its candidate target sites. We applied a series of comparative experiments to demonstrate that TEC-miTarget significantly improves microRNA target prediction, compared with existing state-of-the-art models. Our approach is the first approach to perform comparisons with other approaches at both sequence and transcript levels. Furthermore, it is the first approach compared with both deep learning-based and seed-match-based methods. We first compared TEC-miTarget's performance with approaches at the sequence level, and our approach delivers substantial improvements in performance using the same datasets and evaluation metrics. Moreover, we utilized TEC-miTarget to predict microRNA targets in long mRNA sequences, which involves two steps: selecting candidate target site sequences and applying sequence-level predictions. We finally showed that TEC-miTarget outperforms other approaches at the transcript level, including the popular seed match methods widely used in previous years. CONCLUSIONS: We propose a novel approach for predicting microRNA targets at both sequence and transcript levels, and demonstrate that our approach outperforms other methods based on deep learning or seed match. We also provide our approach as an easy-to-use software, TEC-miTarget, at https://github.com/tingpeng17/TEC-miTarget . Our results provide new perspectives for microRNA target prediction.

Enhancing assisted diagnostic accuracy in scalp psoriasis: A Multi-Network Fusion Object Detection Framework for dermoscopic pattern diagnosis.

BACKGROUND: Dermoscopy is a common method of scalp psoriasis diagnosis, and several artificial intelligence techniques have been used to assist dermoscopy in the diagnosis of nail fungus disease, the most commonly used being the convolutional neural network algorithm; however, convolutional neural networks are only the most basic algorithm, and the use of object detection algorithms to assist dermoscopy in the diagnosis of scalp psoriasis has not been reported. OBJECTIVES: Establishment of a dermoscopic modality diagnostic framework for scalp psoriasis based on object detection technology and image enhancement to improve diagnostic efficiency and accuracy. METHODS: We analyzed the dermoscopic patterns of scalp psoriasis diagnosed at 72nd Group army hospital of PLA from January 1, 2020 to December 31, 2021, and selected scalp seborrheic dermatitis as a control group. Based on dermoscopic images and major dermoscopic patterns of scalp psoriasis and scalp seborrheic dermatitis, we investigated a multi-network fusion object detection framework based on the object detection technique Faster R-CNN and the image enhancement technique contrast limited adaptive histogram equalization (CLAHE), for assisting in the diagnosis of scalp psoriasis and scalp seborrheic dermatitis, as well as to differentiate the major dermoscopic patterns of the two diseases. The diagnostic performance of the multi-network fusion object detection framework was compared with that between dermatologists. RESULTS: A total of 1876 dermoscopic images were collected, including 1218 for scalp psoriasis versus 658 for scalp seborrheic dermatitis. Based on these images, training and testing are performed using a multi-network fusion object detection framework. The results showed that the test accuracy, specificity, sensitivity, and Youden index for the diagnosis of scalp psoriasis was: 91.0%, 89.5%, 91.0%, and 0.805, and for the main dermoscopic patterns of scalp psoriasis and scalp seborrheic dermatitis, the diagnostic results were: 89.9%, 97.7%, 89.9%, and 0.876. Comparing the diagnostic results with those of five dermatologists, the fusion framework performs better than the dermatologists' diagnoses. CONCLUSIONS: Studies have shown some differences in dermoscopic patterns between scalp psoriasis and scalp seborrheic dermatitis. The proposed multi-network fusion object detection framework has higher diagnostic performance for scalp psoriasis than for dermatologists.

Design of a novel intelligent computing framework for predictive solutions of malaria propagation model.

The paper presents an innovative computational framework for predictive solutions for simulating the spread of malaria. The structure incorporates sophisticated computing methods to improve the reliability of predicting malaria outbreaks. The study strives to provide a strong and effective tool for forecasting the propagation of malaria via the use of an AI-based recurrent neural network (RNN). The model is classified into two groups, consisting of humans and mosquitoes. To develop the model, the traditional Ross-Macdonald model is expanded upon, allowing for a more comprehensive analysis of the intricate dynamics at play. To gain a deeper understanding of the extended Ross model, we employ RNN, treating it as an initial value problem involving a system of first-order ordinary differential equations, each representing one of the seven profiles. This method enables us to obtain valuable insights and elucidate the complexities inherent in the propagation of malaria. Mosquitoes and humans constitute the two cohorts encompassed within the exposition of the mathematical dynamical model. Human dynamics are comprised of individuals who are susceptible, exposed, infectious, and in recovery. The mosquito population, on the other hand, is divided into three categories: susceptible, exposed, and infected. For RNN, we used the input of 0 to 300 days with an interval length of 3 days. The evaluation of the precision and accuracy of the methodology is conducted by superimposing the estimated solution onto the numerical solution. In addition, the outcomes obtained from the RNN are examined, including regression analysis, assessment of error autocorrelation, examination of time series response plots, mean square error, error histogram, and absolute error. A reduced mean square error signifies that the model's estimates are more accurate. The result is consistent with acquiring an approximate absolute error close to zero, revealing the efficacy of the suggested strategy. This research presents a novel approach to solving the malaria propagation model using recurrent neural networks. Additionally, it examines the behavior of various profiles under varying initial conditions of the malaria propagation model, which consists of a system of ordinary differential equations.

A probabilistic knowledge graph for target identification.

Early identification of safe and efficacious disease targets is crucial to alleviating the tremendous cost of drug discovery projects. However, existing experimental methods for identifying new targets are generally labor-intensive and failure-prone. On the other hand, computational approaches, especially machine learning-based frameworks, have shown remarkable application potential in drug discovery. In this work, we propose Progeni, a novel machine learning-based framework for target identification. In addition to fully exploiting the known heterogeneous biological networks from various sources, Progeni integrates literature evidence about the relations between biological entities to construct a probabilistic knowledge graph. Graph neural networks are then employed in Progeni to learn the feature embeddings of biological entities to facilitate the identification of biologically relevant target candidates. A comprehensive evaluation of Progeni demonstrated its superior predictive power over the baseline methods on the target identification task. In addition, our extensive tests showed that Progeni exhibited high robustness to the negative effect of exposure bias, a common phenomenon in recommendation systems, and effectively identified new targets that can be strongly supported by the literature. Moreover, our wet lab experiments successfully validated the biological significance of the top target candidates predicted by Progeni for melanoma and colorectal cancer. All these results suggested that Progeni can identify biologically effective targets and thus provide a powerful and useful tool for advancing the drug discovery process.

A hybrid 1D CNN-BiLSTM model for epileptic seizure detection using multichannel EEG feature fusion.

Epilepsy, a chronic non-communicable disease is characterized by repeated unprovoked seizures, which are transient episodes of abnormal electrical activity in the brain. While Electroencephalography (EEG) is considered as the gold standard for diagnosis in current clinical practice, manual inspection of EEG is time consuming and biased. This paper presents a novel hybrid 1D CNN-Bi LSTM feature fusion model for automatically detecting seizures. The proposed model leverages spatial features extracted by one dimensional convolutional neural network and temporal features extracted by bi directional long short-term memory network. Ictal and inter ictal data is first acquired from the long multichannel EEG record. The acquired data is segmented and labelled using small fixed windows. Signal features are then extracted from the segments concurrently by the parallel combination of CNN and Bi-LSTM. The spatial and temporal features thus captured are then fused to enhance classification accuracy of model. The approach is validated using benchmark CHB-MIT dataset and 5-fold cross validation which resulted in an average accuracy of 95.90%, with precision 94.78%, F1 score 95.95%. Notably model achieved average sensitivity of 97.18% with false positivity rate at 0.05/hr. The significantly lower false positivity and false negativity rates indicate that the proposed model is a promising tool for detecting seizures in epilepsy patients. The employed parallel path network benefits from memory function of Bi-LSTM and strong feature extraction capabilities of CNN. Moreover, eliminating the need for any domain transformation or additional preprocessing steps, model effectively reduces complexity and enhances efficiency, making it suitable for use by clinicians during the epilepsy diagnostic process.

Modeling Zinc Complexes Using Neural Networks.

Understanding the energetic landscapes of large molecules is necessary for the study of chemical and biological systems. Recently, deep learning has greatly accelerated the development of models based on quantum chemistry, making it possible to build potential energy surfaces and explore chemical space. However, most of this work has focused on organic molecules due to the simplicity of their electronic structures as well as the availability of data sets. In this work, we build a deep learning architecture to model the energetics of zinc organometallic complexes. To achieve this, we have compiled a configurationally and conformationally diverse data set of zinc complexes using metadynamics to overcome the limitations of traditional sampling methods. In terms of the neural network potentials, our results indicate that for zinc complexes, partial charges play an important role in modeling the long-range interactions with a neural network. Our developed model outperforms semiempirical methods in predicting the relative energy of zinc conformers, yielding a mean absolute error (MAE) of 1.32 kcal/mol with reference to the double-hybrid PWPB95 method.

Segmentation of bone surface from ultrasound using a lightweight network UBS-Net.

Objective. Ultrasound-assisted orthopaedic navigation held promise due to its non-ionizing feature, portability, low cost, and real-time performance. To facilitate the applications, it was critical to have accurate and real-time bone surface segmentation. Nevertheless, the imaging artifacts and low signal-to-noise ratios in the tomographical B-mode ultrasound (B-US) images created substantial challenges in bone surface detection. In this study, we presented an end-to-end lightweight US bone segmentation network (UBS-Net) for bone surface detection.Approach. We presented an end-to-end lightweight UBS-Net for bone surface detection, using the U-Net structure as the base framework and a level set loss function for improved sensitivity to bone surface detectability. A dual attention (DA) mechanism was introduced at the end of the encoder, which considered both position and channel information to obtain the correlation between the position and channel dimensions of the feature map, where axial attention (AA) replaced the traditional self-attention (SA) mechanism in the position attention module for better computational efficiency. The position attention and channel attention (CA) were combined with a two-class fusion module for the DA map. The decoding module finally completed the bone surface detection.Main Results. As a result, a frame rate of 21 frames per second (fps) in detection were achieved. It outperformed the state-of-the-art method with higher segmentation accuracy (Dice similarity coefficient: 88.76% versus 87.22%) when applied the retrospective ultrasound (US) data from 11 volunteers.Significance. The proposed UBS-Net for bone surface detection in ultrasound achieved outstanding accuracy and real-time performance. The new method out-performed the state-of-the-art methods. It had potential in US-guided orthopaedic surgery applications.

Multi-scale feature aggregation and fusion network with self-supervised multi-level perceptual loss for textures preserving low-dose CT denoising.

Objective. The textures and detailed structures in computed tomography (CT) images are highly desirable for clinical diagnosis. This study aims to expand the current body of work on textures and details preserving convolutional neural networks for low-dose CT (LDCT) image denoising task.Approach. This study proposed a novel multi-scale feature aggregation and fusion network (MFAF-net) for LDCT image denoising. Specifically, we proposed a multi-scale residual feature aggregation module to characterize multi-scale structural information in CT images, which captures regional-specific inter-scale variations using learned weights. We further proposed a cross-level feature fusion module to integrate cross-level features, which adaptively weights the contributions of features from encoder to decoder by using a spatial pyramid attention mechanism. Moreover, we proposed a self-supervised multi-level perceptual loss module to generate multi-level auxiliary perceptual supervision for recovery of salient textures and structures of tissues and lesions in CT images, which takes advantage of abundant semantic information at various levels. We introduced parameters for the perceptual loss to adaptively weight the contributions of auxiliary features of different levels and we also introduced an automatic parameter tuning strategy for these parameters.Main results. Extensive experimental studies were performed to validate the effectiveness of the proposed method. Experimental results demonstrate that the proposed method can achieve better performance on both fine textures preservation and noise suppression for CT image denoising task compared with other competitive convolutional neural network (CNN) based methods.Significance. The proposed MFAF-net takes advantage of multi-scale receptive fields, cross-level features integration and self-supervised multi-level perceptual loss, enabling more effective recovering of fine textures and detailed structures of tissues and lesions in CT images.

Construction of a diagnostic model based on random forest and artificial neural network for peri-implantitis.

OBJECTIVES: This study aimed to reveal critical genes regulating peri-implantitis during its development and construct a diagnostic model by using random forest (RF) and artificial neural network (ANN). METHODS: GSE-33774, GSE106090, and GSE57631 datasets were obtained from the GEO database. The GSE33774 and GSE106090 datasets were analyzed for differential expression and functional enrichment. The protein-protein interaction networks (PPI) and RF screened vital genes. A diagnostic model for peri-implantitis was established using ANN and validated on the GSE33774 and GSE57631 datasets. A transcription factor-gene interaction network and a transcription factor-micro-RNA (miRNA) regulatory network were also established. RESULTS: A total of 124 differentially expressed genes (DEGs) involved in the regulation of peri-implantitis were screened. Enrichment analysis showed that DEGs were mainly associated with immune receptor activity and cytokine receptor activity and were mainly involved in processes such as leukocyte and neutrophil migration. The PPI and RF screened six essential genes, namely, CD38, CYBB, FCGR2A, SELL, TLR4, and CXCL8. The receiver operating characteristic curve (ROC) indicated that the ANN model had an excellent diagnostic performance. FOXC1, GATA2, and NF-κB1 may be essential transcription factors in peri-implantitis, and hsa-miR-204 may be a key miRNA. CONCLUSIONS: The diagnostic model of peri-implantitis constructed by RF and ANN has high confidence, and CD38, CYBB, FCGR2A, SELL, TLR4, and CXCL8 are potential diagnostic markers. FOXC1, GATA2, and NF-κB1 may be essential transcription factors in peri-implantitis, and hsa-miR-204 plays a vital role as a critical miRNA.