Analyzing pain patterns in the emergency department: Leveraging clinical text deep learning models for real-world insights.

OBJECTIVE: To determine the incidence of patients presenting in pain to a large Australian inner-city emergency department (ED) using a clinical text deep learning algorithm. MATERIALS AND METHODS: A fine-tuned, domain-specific, transformer-based clinical text deep learning model was used to interpret free-text nursing assessments in the electronic medical records of 235,789 adult presentations to the ED over a three-year period. The model classified presentations according to whether the patient had pain on arrival at the ED. Interrupted time series analysis was used to determine the incidence of pain in patients on arrival over time. We described the changes in the population characteristics and incidence of patients with pain on arrival occurring with the start of the Covid-19 pandemic. RESULTS: 55.16% (95%CI 54.95%-55.36%) of all patients presenting to this ED had pain on arrival. There were differences in demographics and arrival and departure patterns between patients with and without pain. The Covid-19 pandemic initially precipitated a decrease followed by a sharp, sustained rise in pain on arrival, with concurrent changes to the population arriving in pain and their treatment. DISCUSSION: Applying a clinical text deep learning model has successfully identified the incidence of pain on arrival. It represents an automated, reproducible mechanism to identify pain from routinely collected medical records. The description of this population and their treatment forms the basis of intervention to improve care for patients with pain. The combination of the clinical text deep learning models and interrupted time series analysis has reported on the effects of the Covid-19 pandemic on pain care in the ED, outlining a methodology to assess the impact of significant events or interventions on pain care in the ED. CONCLUSION: Applying a novel deep learning approach to identifying pain guides methodological approaches to evaluating pain care interventions in the ED, giving previously unavailable population-level insights.

Interpretation of Artificial Intelligence Models in Healthcare: A Pictorial Guide for Clinicians.

Artificial intelligence (AI) models can play a more effective role in managing patients with the explosion of digital health records available in the healthcare industry. Machine-learning (ML) and deep-learning (DL) techniques are two methods used to develop predictive models that serve to improve the clinical processes in the healthcare industry. These models are also implemented in medical imaging machines to empower them with an intelligent decision system to aid physicians in their decisions and increase the efficiency of their routine clinical practices. The physicians who are going to work with these machines need to have an insight into what happens in the background of the implemented models and how they work. More importantly, they need to be able to interpret their predictions, assess their performance, and compare them to find the one with the best performance and fewer errors. This review aims to provide an accessible overview of key evaluation metrics for physicians without AI expertise. In this review, we developed four real-world diagnostic AI models (two ML and two DL models) for breast cancer diagnosis using ultrasound images. Then, 23 of the most commonly used evaluation metrics were reviewed uncomplicatedly for physicians. Finally, all metrics were calculated and used practically to interpret and evaluate the outputs of the models. Accessible explanations and practical applications empower physicians to effectively interpret, evaluate, and optimize AI models to ensure safety and efficacy when integrated into clinical practice.

An ensemble model for accurate prediction of key water quality parameters in river based on deep learning methods.

Deep learning models provide a more powerful method for accurate and stable prediction of water quality in rivers, which is crucial for the intelligent management and control of the water environment. To increase the accuracy of predicting the water quality parameters and learn more about the impact of complex spatial information based on deep learning models, this study proposes two ensemble models TNX (with temporal attention) and STNX (with spatio-temporal attention) based on seasonal and trend decomposition (STL) method to predict water quality using geo-sensory time series data. Dissolved oxygen, total phosphorus, and ammonia nitrogen were predicted in short-step (1 h, and 2 h) and long-step (12 h, and 24 h) with seven water quality monitoring sites in a river. The ensemble model TNX improved the performance by 2.1%-6.1% and 4.3%-22.0% relative to the best baseline deep learning model for the short-step and long-step water quality prediction, and it can capture the variation pattern of water quality parameters by only predicting the trend component of raw data after STL decomposition. The STNX model, with spatio-temporal attention, obtained 0.5%-2.4% and 2.3%-5.7% higher performance compared to the TNX model for the short-step and long-step water quality prediction, and such improvement was more effective in mitigating the prediction shift patterns of long-step prediction. Moreover, the model interpretation results consistently demonstrated positive relationship patterns across all monitoring sites. However, the significance of seven specific monitoring sites diminished as the distance between the predicted and input monitoring sites increased. This study provides an ensemble modeling approach based on STL decomposition for improving short-step and long-step prediction of river water quality parameter, and understands the impact of complex spatial information on deep learning model.

Leveraging deep learning for detecting red blood cell morphological changes in blood films from children with severe malaria anaemia.

In sub-Saharan Africa, acute-onset severe malaria anaemia (SMA) is a critical challenge, particularly affecting children under five. The acute drop in haematocrit in SMA is thought to be driven by an increased phagocytotic pathological process in the spleen, leading to the presence of distinct red blood cells (RBCs) with altered morphological characteristics. We hypothesized that these RBCs could be detected systematically and at scale in peripheral blood films (PBFs) by harnessing the capabilities of deep learning models. Assessment of PBFs by a microscopist does not scale for this task and is subject to variability. Here we introduce a deep learning model, leveraging a weakly supervised Multiple Instance Learning framework, to Identify SMA (MILISMA) through the presence of morphologically changed RBCs. MILISMA achieved a classification accuracy of 83% (receiver operating characteristic area under the curve [AUC] of 87%; precision-recall AUC of 76%). More importantly, MILISMA's capabilities extend to identifying statistically significant morphological distinctions (p < 0.01) in RBCs descriptors. Our findings are enriched by visual analyses, which underscore the unique morphological features of SMA-affected RBCs when compared to non-SMA cells. This model aided detection and characterization of RBC alterations could enhance the understanding of SMA's pathology and refine SMA diagnostic and prognostic evaluation processes at scale.

Advanced deep learning algorithm for instant discriminating of tea leave stress symptoms by smartphone-based detection.

The primary challenges in tea production under multiple stress exposures have negatively affected its global market sustainability, so introducing an infield fast technique for monitoring tea leaves' stresses has tremendous urgent needs. Therefore, this study aimed to propose an efficient method for the detection of stress symptoms based on a portable smartphone with deep learning models. Firstly, a database containing over 10,000 images of tea garden canopies in complex natural scenes was developed, which included healthy (no stress) and three types of stress (tea anthracnose (TA), tea blister blight (TB) and sunburn (SB)). Then, YOLOv5m and YOLOv8m algorithms were adapted to discriminate the four types of stress symptoms; where the YOLOv8m algorithm achieved better performance in the identification of healthy leaves (98%), TA (92.0%), TB (68.4%) and SB (75.5%). Furthermore, the YOLOv8m algorithm was used to construct a model for differentiation of disease severity of TA, and a satisfactory result was obtained with the accuracy of mild, moderate, and severe TA infections were 94%, 96%, and 91%, respectively. Besides, we found that CNN kernels of YOLOv8m could efficiently extract the texture characteristics of the images at layer 2, and these characteristics can clearly distinguish different types of stress symptoms. This makes great contributions to the YOLOv8m model to achieve high-precision differentiation of four types of stress symptoms. In conclusion, our study provided an effective system to achieve low-cost, high-precision, fast, and infield diagnosis of tea stress symptoms in complex natural scenes based on smartphone and deep learning algorithms.

AI-Driven Clinical Decision Support Systems: An Ongoing Pursuit of Potential.

Clinical Decision Support Systems (CDSS) are essential tools in contemporary healthcare, enhancing clinicians' decisions and patient outcomes. The integration of artificial intelligence (AI) is now revolutionizing CDSS even further. This review delves into AI technologies transforming CDSS, their applications in healthcare decision-making, associated challenges, and the potential trajectory toward fully realizing AI-CDSS's potential. The review begins by laying the groundwork with a definition of CDSS and its function within the healthcare field. It then highlights the increasingly significant role that AI is playing in enhancing CDSS effectiveness and efficiency, underlining its evolving prominence in shaping healthcare practices. It examines the integration of AI technologies into CDSS, including machine learning algorithms like neural networks and decision trees, natural language processing, and deep learning. It also addresses the challenges associated with AI integration, such as interpretability and bias. We then shift to AI applications within CDSS, with real-life examples of AI-driven diagnostics, personalized treatment recommendations, risk prediction, early intervention, and AI-assisted clinical documentation. The review emphasizes user-centered design in AI-CDSS integration, addressing usability, trust, workflow, and ethical and legal considerations. It acknowledges prevailing obstacles and suggests strategies for successful AI-CDSS adoption, highlighting the need for workflow alignment and interdisciplinary collaboration. The review concludes by summarizing key findings, underscoring AI's transformative potential in CDSS, and advocating for continued research and innovation. It emphasizes the need for collaborative efforts to realize a future where AI-powered CDSS optimizes healthcare delivery and improves patient outcomes.

Unlocking the power of AI models: exploring protein folding prediction through comparative analysis.

Protein structure determination has made progress with the aid of deep learning models, enabling the prediction of protein folding from protein sequences. However, obtaining accurate predictions becomes essential in certain cases where the protein structure remains undescribed. This is particularly challenging when dealing with rare, diverse structures and complex sample preparation. Different metrics assess prediction reliability and offer insights into result strength, providing a comprehensive understanding of protein structure by combining different models. In a previous study, two proteins named ARM58 and ARM56 were investigated. These proteins contain four domains of unknown function and are present in Leishmania spp. ARM refers to an antimony resistance marker. The study's main objective is to assess the accuracy of the model's predictions, thereby providing insights into the complexities and supporting metrics underlying these findings. The analysis also extends to the comparison of predictions obtained from other species and organisms. Notably, one of these proteins shares an ortholog with Trypanosoma cruzi and Trypanosoma brucei, leading further significance to our analysis. This attempt underscored the importance of evaluating the diverse outputs from deep learning models, facilitating comparisons across different organisms and proteins. This becomes particularly pertinent in cases where no previous structural information is available.

Synergistic integration of deep learning with protein docking in cardiovascular disease treatment strategies.

This research delves into the exploration of the potential of tocopherol-based nanoemulsion as a therapeutic agent for cardiovascular diseases (CVD) through an in-depth molecular docking analysis. The study focuses on elucidating the molecular interactions between tocopherol and seven key proteins (1O8a, 4YAY, 4DLI, 1HW9, 2YCW, 1BO9 and 1CX2) that play pivotal roles in CVD development. Through rigorous in silico docking investigations, assessment was conducted on the binding affinities, inhibitory potentials and interaction patterns of tocopherol with these target proteins. The findings revealed significant interactions, particularly with 4YAY, displaying a robust binding energy of -6.39 kcal/mol and a promising Ki value of 20.84 µM. Notable interactions were also observed with 1HW9, 4DLI, 2YCW and 1CX2, further indicating tocopherol's potential therapeutic relevance. In contrast, no interaction was observed with 1BO9. Furthermore, an examination of the common residues of 4YAY bound to tocopherol was carried out, highlighting key intermolecular hydrophobic bonds that contribute to the interaction's stability. Tocopherol complies with pharmacokinetics (Lipinski's and Veber's) rules for oral bioavailability and proves safety non-toxic and non-carcinogenic. Thus, deep learning-based protein language models ESM1-b and ProtT5 were leveraged for input encodings to predict interaction sites between the 4YAY protein and tocopherol. Hence, highly accurate predictions of these critical protein-ligand interactions were achieved. This study not only advances the understanding of these interactions but also highlights deep learning's immense potential in molecular biology and drug discovery. It underscores tocopherol's promise as a cardiovascular disease management candidate, shedding light on its molecular interactions and compatibility with biomolecule-like characteristics.

Deep learning models for atypical serotonergic cells recognition.

BACKGROUND: The serotonergic system modulates brain processes via functionally distinct subpopulations of neurons with heterogeneous properties, including their electrophysiological activity. In extracellular recordings, serotonergic neurons to be investigated for their functional properties are commonly identified on the basis of "typical" features of their activity, i.e. slow regular firing and relatively long duration of action potentials. Thus, due to the lack of equally robust criteria for discriminating serotonergic neurons with "atypical" features from non-serotonergic cells, the physiological relevance of the diversity of serotonergic neuron activities results largely understudied. NEW METHODS: We propose deep learning models capable of discriminating typical and atypical serotonergic neurons from non-serotonergic cells with high accuracy. The research utilized electrophysiological in vitro recordings from serotonergic neurons identified by the expression of fluorescent proteins specific to the serotonergic system and non-serotonergic cells. These recordings formed the basis of the training, validation, and testing data for the deep learning models. The study employed convolutional neural networks (CNNs), known for their efficiency in pattern recognition, to classify neurons based on the specific characteristics of their action potentials. RESULTS: The models were trained on a dataset comprising 27,108 original action potential samples, alongside an extensive set of 12 million synthetic action potential samples, designed to mitigate the risk of overfitting the background noise in the recordings, a potential source of bias. Results show that the models achieved high accuracy and were further validated on "non-homogeneous" data, i.e., data unknown to the model and collected on different days from those used for the training of the model, to confirm their robustness and reliability in real-world experimental conditions. COMPARISON WITH EXISTING METHODS: Conventional methods for identifying serotonergic neurons allow recognition of serotonergic neurons defined as typical. Our model based on the analysis of the sole action potential reliably recognizes over 94% of serotonergic neurons including those with atypical features of spike and activity. CONCLUSION: The model is ready for use in experiments conducted with the here described recording parameters. We release the codes and procedures allowing to adapt the model to different acquisition parameters or for identification of other classes of spontaneously active neurons.

A deep learning based multi-model approach for predicting drug-like chemical compound's toxicity.

Ensuring the safety and efficacy of chemical compounds is crucial in small-molecule drug development. In the later stages of drug development, toxic compounds pose a significant challenge, losing valuable resources and time. Early and accurate prediction of compound toxicity using deep learning models offers a promising solution to mitigate these risks during drug discovery. In this study, we present the development of several deep-learning models aimed at evaluating different types of compound toxicity, including acute toxicity, carcinogenicity, hERG_cardiotoxicity (the human ether-a-go-go related gene caused cardiotoxicity), hepatotoxicity, and mutagenicity. To address the inherent variations in data size, label type, and distribution across different types of toxicity, we employed diverse training strategies. Our first approach involved utilizing a graph convolutional network (GCN) regression model to predict acute toxicity, which achieved notable performance with Pearson R 0.76, 0.74, and 0.65 for intraperitoneal, intravenous, and oral administration routes, respectively. Furthermore, we trained multiple GCN binary classification models, each tailored to a specific type of toxicity. These models exhibited high area under the curve (AUC) scores, with an impressive AUC of 0.69, 0.77, 0.88, and 0.79 for predicting carcinogenicity, hERG_cardiotoxicity, mutagenicity, and hepatotoxicity, respectively. Additionally, we have used the approved drug dataset to determine the appropriate threshold value for the prediction score in model usage. We integrated these models into a virtual screening pipeline to assess their effectiveness in identifying potential low-toxicity drug candidates. Our findings indicate that this deep learning approach has the potential to significantly reduce the cost and risk associated with drug development by expediting the selection of compounds with low toxicity profiles. Therefore, the models developed in this study hold promise as critical tools for early drug candidate screening and selection.

A hands-on guide to use network video recorders, internet protocol cameras, and deep learning models for dynamic monitoring of trout and salmon in small streams.

This study outlines a method for using surveillance cameras and an algorithm that calls a deep learning model to generate video segments featuring salmon and trout in small streams. This automated process greatly reduces the need for human intervention in video surveillance. Furthermore, a comprehensive guide is provided on setting up and configuring surveillance equipment, along with instructions on training a deep learning model tailored to specific requirements. Access to video data and knowledge about deep learning models makes monitoring of trout and salmon dynamic and hands-on, as the collected data can be used to train and further improve deep learning models. Hopefully, this setup will encourage fisheries managers to conduct more monitoring as the equipment is relatively cheap compared with customized solutions for fish monitoring. To make effective use of the data, natural markings of the camera-captured fish can be used for individual identification. While the automated process greatly reduces the need for human intervention in video surveillance and speeds up the initial sorting and detection of fish, the manual identification of individual fish based on natural markings still requires human effort and involvement. Individual encounter data hold many potential applications, such as capture-recapture and relative abundance models, and for evaluating fish passages in streams with hydropower by spatial recaptures, that is, the same individual identified at different locations. There is much to gain by using this technique as camera captures are the better option for the fish's welfare and are less time-consuming compared with physical captures and tagging.

Prediction of vitreomacular traction syndrome outcomes with deep learning: A pilot study.

PURPOSE: To investigate the potential of an Optical Coherence Tomography (OCT) based Deep-Learning (DL) model in the prediction of Vitreomacular Traction (VMT) syndrome outcomes. DESIGN: A single-centre retrospective review. METHODS: Records of consecutive adult patients attending the Royal Adelaide Hospital vitreoretinal clinic with evidence of spontaneous VMT were reviewed from January 2019 until May 2022. All patients with evidence of causes of cystoid macular oedema or secondary causes of VMT were excluded. OCT scans and outcome data obtained from patient records was used to train, test and then validate the models. RESULTS: For the deep learning model, ninety-five patient files were identified from the OCT (SPECTRALIS system; Heidelberg Engineering, Heidelberg, Germany) records. 25% of the patients spontaneously improved, 48% remained stable and 27% had progression of their disease, approximately. The final longitudinal model was able to predict 'improved' or 'stable' disease with a positive predictive value of 0.72 and 0.79, respectively. The accuracy of the model was greater than 50%. CONCLUSIONS: Deep-learning models may be utilised in real-world settings to predict outcomes of VMT. This approach requires further investigation as it may improve patient outcomes by aiding ophthalmologists in cross-checking management decisions and reduce the need for unnecessary interventions or delays.

Development of artificial intelligence edge computing based wearable device for fall detection and prevention of elderly people.

Elderly falls are a major concerning threat resulting in over 1.5-2 million elderly people experiencing severe injuries and 1 million deaths yearly. Falls experienced by Elderly people may lead to a long-term negative impact on their physical and psychological health conditions. Major healthcare research had focused on this lately to detect and prevent the fall. In this work, an Artificial Intelligence (AI) edge computing based wearable device is designed and developed for detection and prevention of fall of elderly people. Further, the various deep learning algorithms such as Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU) are utilized for activity recognition of elderly. Also, the CNN-LSTM, RNN-LSTM and GRU-LSTM with and without attention layer respectively are utilized and the performance metrics are analyzed to find the best deep learning model. Furthermore, the three different hardware boards such as Jetson Nano developer board, Raspberry PI 3 and 4 are utilized as an AI edge computing device and the best deep learning model is implemented and the computation time is evaluated. Results demonstrate that the CNN-LSTM with attention layer exhibits the accuracy, recall, precision and F1_Score of 97%, 98%, 98% and 0.98 respectively which is better when compared to other deep learning models. Also, the computation time of NVIDIA Jetson Nano is less when compared to other edge computing devices. This work appears to be of high societal relevance since the proposed wearable device can be used to monitor the activity of elderly and prevents the elderly falls which improve the quality of life of elderly people.

Comparative analysis of the spatial distribution of brain metastases across several primary cancers using machine learning and deep learning models.

OBJECTIVE: Brain metastases (BM) are associated with poor prognosis and increased mortality rates, making them a significant clinical challenge. Studying BMs can aid in improving early detection and monitoring. Systematic comparisons of anatomical distributions of BM from different primary cancers, however, remain largely unavailable. METHODS: To test the hypothesis that anatomical BM distributions differ based on primary cancer type, we analyze the spatial coordinates of BMs for five different primary cancer types along principal component (PC) axes. The dataset includes 3949 intracranial metastases, labeled by primary cancer types and with six features. We employ PC coordinates to highlight the distinctions between various cancer types. We utilized different Machine Learning (ML) algorithms (RF, SVM, TabNet DL) models to establish the relationship between primary cancer diagnosis, spatial coordinates of BMs, age, and target volume. RESULTS: Our findings revealed that PC1 aligns most with the Y axis, followed by the Z axis, and has minimal correlation with the X axis. Based on PC1 versus PC2 plots, we identified notable differences in anatomical spreading patterns between Breast and Lung cancer, as well as Breast and Renal cancer. In contrast, Renal and Lung cancer, as well as Lung and Melanoma, showed similar patterns. Our ML and DL results demonstrated high accuracy in distinguishing BM distribution for different primary cancers, with the SVM algorithm achieving 97% accuracy using a polynomial kernel and TabNet achieving 96%. The RF algorithm ranked PC1 as the most important discriminating feature. CONCLUSIONS: In summary, our results support accurate multiclass ML classification regarding brain metastases distribution.

Using an interpretable deep learning model for the prediction of riverine suspended sediment load.

The prediction of suspended sediment load (SSL) within riverine systems is critical to understanding the watershed's hydrology. Therefore, the novelty of our research is developing an interpretable (explainable) model based on deep learning (DL) and Shapley Additive ExPlanations (SHAP) interpretation technique for prediction of SSL in the riverine systems. This paper investigates the abilities of four DL models, including dense deep neural networks (DDNN), long short-term memory (LSTM), gated recurrent unit (GRU), and simple recurrent neural network (RNN) models for the prediction of daily SSL using river discharge and rainfall data at a daily time scale in the Taleghan River watershed, northwestern Tehran, Iran. The performance of models was evaluated by using several quantitative and graphical criteria. The effect of parameter settings on the performance of deep models on SSL prediction was also investigated. The optimal optimization algorithms, maximum iteration (MI), and batch size (BC) were obtained for modeling daily SSL, and structure of the model impact on prediction remarkably. The comparison of prediction accuracy of the models illustrated that DDNN (with R2 = 0.96, RMSE = 333.46) outperformed LSTM (R2 = 0.75, RMSE = 786.20), GRU (R2 = 0.73, RMSE = 825.67), and simple RNN (R2 = 0.78, RMSE = 741.45). Furthermore, the Taylor diagram confirmed that DDNN has the highest performance among other models. Interpretation techniques can address the black-box nature of models, and here, SHAP was applied to develop an interpretable DL model to interpret of DL model's output. The results of SHAP showed that river discharge has the strongest impact on the model's output in estimating SSL. Overall, we conclude that DL models have great potential in watersheds to predict SSL. Therefore, different interpretation techniques as tools to interpret DL model's output (DL model is as black-box model) are recommended in future research.

Analyzing Data Modalities for Cattle Weight Estimation Using Deep Learning Models.

We investigate the impact of different data modalities for cattle weight estimation. For this purpose, we collect and present our own cattle dataset representing the data modalities: RGB, depth, combined RGB and depth, segmentation, and combined segmentation and depth information. We explore a recent vision-transformer-based zero-shot model proposed by Meta AI Research for producing the segmentation data modality and for extracting the cattle-only region from the images. For experimental analysis, we consider three baseline deep learning models. The objective is to assess how the integration of diverse data sources influences the accuracy and robustness of the deep learning models considering four different performance metrics: mean absolute error (MAE), root mean squared error (RMSE), mean absolute percentage error (MAPE), and R-squared (R2). We explore the synergies and challenges associated with each modality and their combined use in enhancing the precision of cattle weight prediction. Through comprehensive experimentation and evaluation, we aim to provide insights into the effectiveness of different data modalities in improving the performance of established deep learning models, facilitating informed decision-making for precision livestock management systems.

Screening and staging of chronic obstructive pulmonary disease with deep learning based on chest X-ray images and clinical parameters.

BACKGROUND: Chronic obstructive pulmonary disease (COPD) is underdiagnosed with the current gold standard measure pulmonary function test (PFT). A more sensitive and simple option for early detection and severity evaluation of COPD could benefit practitioners and patients. METHODS: In this multicenter retrospective study, frontal chest X-ray (CXR) images and related clinical information of 1055 participants were collected and processed. Different deep learning algorithms and transfer learning models were trained to classify COPD based on clinical data and CXR images from 666 subjects, and validated in internal test set based on 284 participants. External test including 105 participants was also performed to verify the generalization ability of the learning algorithms in diagnosing COPD. Meanwhile, the model was further used to evaluate disease severity of COPD by predicting different grads. RESULTS: The Ensemble model showed an AUC of 0.969 in distinguishing COPD by simultaneously extracting fusion features of clinical parameters and CXR images in internal test, better than models that used clinical parameters (AUC = 0.963) or images (AUC = 0.946) only. For the external test set, the AUC slightly declined to 0.934 in predicting COPD based on clinical parameters and CXR images. When applying the Ensemble model to determine disease severity of COPD, the AUC reached 0.894 for three-classification and 0.852 for five-classification respectively. CONCLUSION: The present study used DL algorithms to screen COPD and predict disease severity based on CXR imaging and clinical parameters. The models showed good performance and the approach might be an effective case-finding tool with low radiation dose for COPD diagnosis and staging.

A deep learning-based framework (Co-ReTr) for auto-segmentation of non-small cell-lung cancer in computed tomography images.

PURPOSE: Deep learning-based auto-segmentation algorithms can improve clinical workflow by defining accurate regions of interest while reducing manual labor. Over the past decade, convolutional neural networks (CNNs) have become prominent in medical image segmentation applications. However, CNNs have limitations in learning long-range spatial dependencies due to the locality of the convolutional layers. Transformers were introduced to address this challenge. In transformers with self-attention mechanism, even the first layer of information processing makes connections between distant image locations. Our paper presents a novel framework that bridges these two unique techniques, CNNs and transformers, to segment the gross tumor volume (GTV) accurately and efficiently in computed tomography (CT) images of non-small cell-lung cancer (NSCLC) patients. METHODS: Under this framework, input of multiple resolution images was used with multi-depth backbones to retain the benefits of high-resolution and low-resolution images in the deep learning architecture. Furthermore, a deformable transformer was utilized to learn the long-range dependency on the extracted features. To reduce computational complexity and to efficiently process multi-scale, multi-depth, high-resolution 3D images, this transformer pays attention to small key positions, which were identified by a self-attention mechanism. We evaluated the performance of the proposed framework on a NSCLC dataset which contains 563 training images and 113 test images. Our novel deep learning algorithm was benchmarked against five other similar deep learning models. RESULTS: The experimental results indicate that our proposed framework outperforms other CNN-based, transformer-based, and hybrid methods in terms of Dice score (0.92) and Hausdorff Distance (1.33). Therefore, our proposed model could potentially improve the efficiency of auto-segmentation of early-stage NSCLC during the clinical workflow. This type of framework may potentially facilitate online adaptive radiotherapy, where an efficient auto-segmentation workflow is required. CONCLUSIONS: Our deep learning framework, based on CNN and transformer, performs auto-segmentation efficiently and could potentially assist clinical radiotherapy workflow.

American Sign Language Recognition and Translation Using Perception Neuron Wearable Inertial Motion Capture System.

Sign language is designed as a natural communication method to convey messages among the deaf community. In the study of sign language recognition through wearable sensors, the data sources are limited, and the data acquisition process is complex. This research aims to collect an American sign language dataset with a wearable inertial motion capture system and realize the recognition and end-to-end translation of sign language sentences with deep learning models. In this work, a dataset consisting of 300 commonly used sentences is gathered from 3 volunteers. In the design of the recognition network, the model mainly consists of three layers: convolutional neural network, bi-directional long short-term memory, and connectionist temporal classification. The model achieves accuracy rates of 99.07% in word-level evaluation and 97.34% in sentence-level evaluation. In the design of the translation network, the encoder-decoder structured model is mainly based on long short-term memory with global attention. The word error rate of end-to-end translation is 16.63%. The proposed method has the potential to recognize more sign language sentences with reliable inertial data from the device.

Evaluating the influence of road construction on landslide susceptibility in Saudi Arabia's mountainous terrain: a Bayesian-optimised deep learning approach with attention mechanism and sensitivity analysis.

In the mountainous region of Asir region of Saudi Arabia, road construction activities are closely associated with frequent landslides, posing significant risks to both human life and infrastructural development. This highlights an urgent need for a highly accurate landslide susceptibility map to guide future development and risk mitigation strategies. Therefore, this study aims to (1) develop robust well-optimised deep learning (DL) models for predicting landslide susceptibility and (2) conduct a comprehensive sensitivity analysis to quantify the impact of each parameter influencing landslides. To achieve these aims, three advanced DL models-Deep Neural Networks (DNN), Convolutional Neural Networks (CNN), and Bayesian-optimised CNN with an attention mechanism-were rigorously trained and validated. Model validation included eight matrices, calibration curves, and Receiver Operating Characteristic (ROC) and Precision-Recall curves. Multicollinearity was examined using Variance Inflation Factor (VIF) to ensure variable independence. Additionally, sensitivity analysis was used to interpret the models and explore the influence of parameters on landslide. Results showed that road networks significantly influenced the areas identified as high-risk zones. Specifically, in the 1-km buffer around roadways, CNN_AM identified 10.42% of the area as 'Very High' susceptibility-more than double the 4.04% indicated by DNN. In the extended 2-km buffer zone around roadways, Bayesian CNN_AM continued to flag a larger area as Very High risk (7.46%), in contrast to DNN's 3.07%. In performance metrics, CNN_AM outshined DNN and regular CNN models, achieving near-perfect scores in Area Under the Curve (AUC), precision-recall, and overall accuracy. Sensitivity analysis highlighted 'Soil Texture', 'Geology', 'Distance to Road', and 'Slope' as crucial for landslide prediction. This research offers a robust, high-accuracy model that emphasises the role of road networks in landslide susceptibility, thereby providing valuable insights for planners and policymakers to proactively mitigate landslide risks in vulnerable zones near existing and future road infrastructure.