J Magn Reson. 2024 Nov 29;370:107812. doi: 10.1016/j.jmr.2024.107812. Online ahead of print.

ABSTRACT

In this study, we introduce a denoising method aimed at improving the contrast ratio in low-field MRI (LFMRI) using an advanced 3D deep convolutional residual network model. Our approach employs synthetic brain imaging datasets that closely mimic the contrast and noise characteristics of LFMRI scans, addressing the limitation of available in-vivo LFMRI datasets for training deep learning models. In the simulation data, the Relative Contrast Ratio (RCR) increased, and similar improvements were observed in the in-vivo data across different imaging conditions. Comparative evaluations demonstrate that our model performs better than the widely used non-deep learning method, BM4D, in enhancing RCR and maintaining high spatial frequency components in in-vivo data.

PMID:39647413 | DOI:10.1016/j.jmr.2024.107812

Comput Methods Programs Biomed. 2024 Dec 4;260:108540. doi: 10.1016/j.cmpb.2024.108540. Online ahead of print.

ABSTRACT

BACKGROUND: Ultrasound (US) is a medical imaging modality that plays a crucial role in the early detection of breast cancer. The emergence of numerous deep learning systems has offered promising avenues for the segmentation and classification of breast cancer tumors in US images. However, challenges such as the absence of data standardization, the exclusion of non-tumor images during training, and the narrow view of single-task methodologies have hindered the practical applicability of these systems, often resulting in biased outcomes. This study aims to explore the potential of multi-task systems in enhancing the detection of breast cancer lesions.

METHODS: To address these limitations, our research introduces an end-to-end multi-task framework designed to leverage the inherent correlations between breast cancer lesion classification and segmentation tasks. Additionally, a comprehensive analysis of a widely utilized public breast cancer ultrasound dataset named BUSI was carried out, identifying its irregularities and devising an algorithm tailored for detecting duplicated images in it.

RESULTS: Experiments are conducted utilizing the curated dataset to minimize potential biases in outcomes. Our multi-task framework exhibits superior performance in breast cancer respecting single-task approaches, achieving improvements close to 15% in segmentation and classification. Moreover, a comparative analysis against the state-of-the-art reveals statistically significant enhancements across both tasks.

CONCLUSION: The experimental findings underscore the efficacy of multi-task techniques, showcasing better generalization capabilities when considering all image types: benign, malignant, and non-tumor images. Consequently, our methodology represents an advance towards more general architectures with real clinical applications in the breast cancer field.

PMID:39647406 | DOI:10.1016/j.cmpb.2024.108540

Food Chem. 2024 Nov 12;467:141999. doi: 10.1016/j.foodchem.2024.141999. Online ahead of print.

ABSTRACT

Timely and effective detection of quality attributes during drying control is essential for enhancing the quality of fruit processing. Consequently, this study aims to employ hyperspectral imaging technology for the non-destructive monitoring of soluble solids content (SSC), titratable acidity (TA), moisture, and hardness in jujubes during hot air drying. Quality parameters were measured at drying temperatures of 55 °C, 60 °C, and 65 °C. A deep learning model (CNN_BiLSTM_SE) was developed, incorporating a convolutioyounal neural network (CNN), bidirectional long short-term memory (BiLSTM), and a squeeze-and-excitation (SE) attention mechanism. The performance of PLSR, SVR, and CNN_BiLSTM_SE was compared using different preprocessing methods (MSC, Baseline, and MSC_1st). The CNN_BiLSTM_SE model, optimized for hyperparameters, outperforms PLSR and SVR in predicting jujube quality attributes. Subsequently, these best prediction models were used to predict quality attributes at the pixel level for jujube, enabling the visualization of the Spatio-temporal distribution of these parameters at different drying stages.

PMID:39647380 | DOI:10.1016/j.foodchem.2024.141999

Comput Biol Med. 2024 Dec 7;185:109503. doi: 10.1016/j.compbiomed.2024.109503. Online ahead of print.

ABSTRACT

GI abnormalities significantly increase mortality rates and impose considerable strain on healthcare systems, underscoring the essential requirement for rapid detection, precise diagnosis, and efficient strategic treatment. To develop a CAD system, this study aims to automatically classify GI disorders utilizing various deep learning methodologies. The proposed system features a three-stage lightweight architecture, consisting of a feature extractor using PSE-CNN, a feature selector employing PCA, and a classifier based on DELM. The framework, designed with only 24 layers and 1.25 million parameters, is employed on the largest dataset, GastroVision, containing 8000 images of 27 GI disorders. To improve visual clarity, a sequential preprocessing strategy is implemented. The model's robustness is evaluated through 5-fold cross-validation. Additionally, several XAI methods, namely Grad-CAM, heatmaps, saliency maps, SHAP, and activation feature maps, are used to explore the model's interpretability. Statistical significance is ensured by calculating the p-value, demonstrating the framework's reliability. The proposed model PSE-CNN-PCA-DELM has achieved outstanding results in the first stage, categorizing the diseases' positions into three primary classes, with average accuracy (97.24 %), precision (97.33 ± 0.01 %), recall (97.24 ± 0.01 %), F1-score (97.33 ± 0.01 %), ROC-AUC (99.38 %), and AUC-PR (98.94 %). In the second stage, the dataset is further divided into nine separate classes, considering the overall disease characteristics, and achieves excellent outcomes with average performance rates of 90.00 %, 89.71 ± 0.11 %, 89.59 ± 0.14 %, 89.51 ± 0.12 %, 98.49 %, and 94.63 %, respectively. The third stage involves a more detailed classification into twenty-seven classes, maintaining strong performance with scores of 93.00 %, 82.69 ± 0.37 %, 83.00 ± 0.38 %, 81.54 ± 0.35 %, 97.38 %, and 88.03 %, respectively. The framework's compact size of 14.88 megabytes and average testing time of 59.17 milliseconds make it highly efficient. Its effectiveness is further validated through comparisons with several TL approaches. Practically, the framework is extremely resilient for clinical implementation.

PMID:39647242 | DOI:10.1016/j.compbiomed.2024.109503

Mol Pharm. 2024 Dec 8. doi: 10.1021/acs.molpharmaceut.4c00946. Online ahead of print.

ABSTRACT

The Biopharmaceutics Classification System (BCS) has facilitated biowaivers and played a significant role in enhancing drug regulation and development efficiency. However, the productivity of measuring the key discriminative properties of BCS, solubility and permeability, still requires improvement, limiting high-throughput applications of BCS, which is essential for evaluating drug candidate developability and guiding formulation decisions in the early stages of drug development. In recent years, advancements in machine learning (ML) and molecular characterization have revealed the potential of quantitative structure-performance relationships (QSPR) for rapid and accurate in silico BCS classification. The present study aims to develop a web platform for high-throughput BCS classification based on high-performance ML models. Initially, four data sets of BCS-related molecular properties: log S, log P, log D, and log Papp were curated. Subsequently, 6 ML algorithms or deep learning frameworks were employed to construct models, with diverse molecular representations ranging from one-dimensional molecular fingerprints, descriptors, and molecular graphs to three-dimensional molecular spatial coordinates. By comparing different combinations of molecular representations and learning algorithms, LightGBM exhibited excellent performance in solubility prediction, with an R2 of 0.84; AttentiveFP outperformed others in permeability prediction, with R2 values of 0.96 and 0.76 for log P and log D, respectively; and XGBoost was the most accurate for log Papp prediction, with an R2 of 0.71. When externally validated on a marketed drug BCS category data set, the best-performing models achieved classification accuracies of over 77 and 73% for solubility and permeability, respectively. Finally, the well-trained models were embedded into the first ML-based BCS class prediction web platform (x f), enabling pharmaceutical scientists to quickly determine the BCS category of candidate drugs, which will aid in the high-throughput BCS assessment for candidate drugs during the preformulation stage, thereby promoting reduced risk and enhanced efficiency in drug development and regulation.

PMID:39647169 | DOI:10.1021/acs.molpharmaceut.4c00946

J Eval Clin Pract. 2025 Feb;31(1):e14253. doi: 10.1111/jep.14253.

ABSTRACT

INTRODUCTION: Understanding students' learning approach, modifying teaching methods, curriculum and material accordingly is essential to deliver quality education. Knowing more about the learning approaches assists in upgrading the profession's quality for continuous professional development.

METHODS: The cross-sectional study was carried out among physiotherapy students studying in physiotherapy colleges affiliated with the same university. The Approaches and Study Skills Inventory for Students questionnaire was used to evaluate learning approaches in both preclinical and clinical students. Data were analysed using the IBM Statistical Package SPSS 27. Statistical significance was set at p < 0.05.

RESULTS: A total of 250 participants with a mean age of 21.09 + 1.93 years, 129 (51.6%) in the preclinical group and 121 (48.4%) in the clinical group participated in the study. 67 (26.7%) of the students were male, while 183 (72.9%) were females. The vast majority of participants (97.6%) adopt a deep approach to learning, while only a small fraction (2.0%) use a surface approach, with the strategic approach being rarely used (0.4%). No significant difference was observed between the males and females, and students of different colleges under the same university.

CONCLUSION: The predominant approach is the deep learning approach reflecting active learning. This may indicate that curriculum and strategies of teaching are employed over physiotherapy students to promote quality learning. Also, the teaching preferences varies between two group of physiotherapy students. Thus, this will also assist physiotherapy educators in planning and delivering learning activities according to learners by knowing their preferences.

PMID:39644511 | DOI:10.1111/jep.14253

Hell J Nucl Med. 2024 Dec 9:s002449912756. doi: 10.1967/s002449912756. Online ahead of print.

ABSTRACT

Positron emission tomography/computed tomography (PET/CT) is a hybrid medical imaging technique that combines PET and CT to provide detailed images of the body's anatomical structures and metabolic activity. It is frequently used for oncology and other medical diagnoses. This overview aims to examine how artificial intelligence (AI) has been used in PET/CT, based on recent state-of-art. There are a number of clinical questions in Nuclear Medicine, and AI could provide answers, having the capability to enhance various aspects of medical imaging. The overview focuses on how machine learning (ML) and deep learning (DL), enhance tumor segmentation, classification, diagnosis, disease-free survival prediction and treatment response prediction in oncology. The analysis showed that the application of AI provides reliable results, especially in the fields of classification and diagnosis. In addition, radiomics is a novel research field enabling quantitative analysis of medical images through feature extraction, utilized for AI model implementation. Despite these advances, addressing issues such as dataset size, standardization, and ethical concerns are essential for broad clinical integration of AI in PET/CT oncology imaging.

PMID:39644273 | DOI:10.1967/s002449912756

J Voice. 2024 Dec 6:S0892-1997(24)00420-X. doi: 10.1016/j.jvoice.2024.11.037. Online ahead of print.

ABSTRACT

Quantification of voice physiology has been a key research goal. Segmenting the glottal area to describe the vocal fold motion has seen increased attention in the last two decades. However, researchers struggled to fully automatize the segmentation task. With the advent of deep learning, fully automated solutions are within reach and have been proposed. Are we then done here? This commentary highlights the open construction sites and how glottis segmentation can be still of scientific interest in this decade.

PMID:39645484 | DOI:10.1016/j.jvoice.2024.11.037

Int Dent J. 2024 Dec 6:S0020-6539(24)01591-0. doi: 10.1016/j.identj.2024.11.005. Online ahead of print.

ABSTRACT

OBJECTIVES: Dental health is integral to overall well-being, with early detection of issues critical for prevention. This research work focuses on utilizing artificial intelligence and deep learning-based object detection techniques for automated detection of common dental issues in orthopantomography x-ray images, including broken roots, periodontally compromised teeth, and the Kennedy classification of partially edentulous arches.

METHODS: An orthopantomography dataset has been used to train several models employing various object detection architectures, hyperparameters, and training techniques. The performance of these models was evaluated to select the one with the highest accuracy. This selected model was subsequently deployed for further testing and validation on unseen data to assess its real-world performance and potential for clinical application.

RESULTS: The proposed model not only facilitates the classification of the Kennedy classification but also offers detailed information about the arch (maxillary or mandibular) and specifies the affected side of the arch (right or left). It can diagnose multiple dental issues simultaneously within an image, enhancing diagnostic capabilities for dental practitioners.

CONCLUSIONS: Despite a small dataset, satisfactory results were achieved through tailored hyperparameters and a piecewise annotation scheme.

PMID:39645471 | DOI:10.1016/j.identj.2024.11.005

BMJ Open. 2024 Aug 31;14(8):e084728. doi: 10.1136/bmjopen-2024-084728.

ABSTRACT

INTRODUCTION: This paper outlines the steps necessary to assess the latest developments in artificial intelligence (AI) as well as Big Data technologies and their relevance to the opioid crisis. Fatal opioid overdoses have risen to over 82 998 annually in the USA. This highlights the need for urgent and effective data-driven solutions. AI approaches, such as machine learning, deep learning and natural language processing, have been employed to analyse patterns and trends in overdose data and facilitate timely interventions. However, a comprehensive scoping review on the effectiveness of AI-driven technologies to detect, treat, prevent or respond to the opioid crisis remains absent. Thus, it is important to identify recent advancements in AI and Big Data technologies in addressing the opioid crisis.

METHODS AND ANALYSIS: We will electronically search four scientific databases (PubMed, Web of Science, Engineering Village and PsycInfo), including finding reference lists and grey literature from 2013 to 2023. Covidence will be used for screening and selecting papers. We will extract information such as citation details, study context, data used, AI/Big Data technologies, features, algorithms and evaluation metrics. This data will be synthesised, analysed and summarised to draw meaningful conclusions and identify future directions to tackle the opioid crisis.

ETHICS AND DISSEMINATION: Ethics approval is not required. Results will be disseminated via conference presentations and peer-reviewed publication.

PMID:39645274 | DOI:10.1136/bmjopen-2024-084728

Radiother Oncol. 2024 Dec 5:110660. doi: 10.1016/j.radonc.2024.110660. Online ahead of print.

ABSTRACT

PURPOSE: To develop and validate a prognostic and predictive model integrating deep learning MRI features and clinical information in patients with stage II nasopharyngeal carcinoma (NPC) to identify patients with a low risk of progression for whom intensity-modulated radiotherapy (IMRT) alone is sufficient.

METHODS: This multicenter, retrospective study enrolled 999 patients with stage II NPC from two centers. 3DResNet was used to extract deep learning MRI features and eXtreme Gradient Boosting model was employed to integrate the pre-trained features and clinical information to obtain an overall score for each patient. Based on the optimal cutoff value of the overall score, patients were stratified into high- and low- risk groups. Model performance was evaluated using concordance indexes (C-indexes), the area under the curve (AUC) values and calibration tests. Survival curves were used to analyze the clinical benefits of additional chemotherapy in each risk group.

RESULTS: The combined model achieved a concordance index of 0.789 (95 % confidence interval [CI] 0.787-0.791), 0.768 (95 % CI 0.764-0.771), and 0.804 (95 % CI 0.801-0.807) for the training, internal validation, and external test cohorts, respectively, demonstrating a statistically significant improvement compared to the MRI model, T Stage, and N Stage. An overall score of < 0.405 in patients was significantly associated with a low risk of progression. In the low-risk group, patients treated with IMRT alone had comparable or even superior progression-free survival (PFS) compared to those who received additional chemotherapy.

CONCLUSION: The model demonstrated a satisfactory prognostic and predictive performance for PFS. Patients with stage II NPC were stratified into different risk groups to help identify optimal candidates who could benefit from IMRT alone.

PMID:39645201 | DOI:10.1016/j.radonc.2024.110660

J Cardiovasc Magn Reson. 2024 Dec 5:101133. doi: 10.1016/j.jocmr.2024.101133. Online ahead of print.

ABSTRACT

BACKGROUND: Pulmonary hypertension (PH) is a heterogeneous condition and regardless of aetiology impacts negatively on survival. Diagnosis of PH is based on hemodynamic parameters measured invasively at right heart catheterization (RHC), however, a non-invasive alternative would be clinically valuable. Our aim was to estimate RHC parameters non-invasively from cardiac MR data using deep learning models and to identify key contributing imaging features.

METHODS: We constructed an explainable convolutional neural network (CNN) taking cardiac MR cine series from 4 different views as input to predict mean pulmonary artery pressure (mPAP). The model was trained and evaluated on 1646 examinations. The model's attention weight and predictive performance associated with each frame, view, or phase was used to judge its importance. Additionally, the importance of each cardiac chamber was inferred by perturbing part of the input pixels.

RESULTS: The model achieved a Pearson Correlation Coefficient (PCC) of 0.80 and R2 of 0.64 in predicting mPAP, and identified the right ventricle (RV) region on short-axis (SAX) view to be especially informative.

CONCLUSIONS: Hemodynamic parameters can be estimated non-invasively with a CNN, using MR cine series from 4 views, revealing key contributing features at the same time.

PMID:39645082 | DOI:10.1016/j.jocmr.2024.101133

Comput Methods Programs Biomed. 2024 Nov 28;260:108537. doi: 10.1016/j.cmpb.2024.108537. Online ahead of print.

ABSTRACT

BACKGROUND: Machine learning-based analysis can accurately detect atrial fibrillation (AF) from photoplethysmograms (PPGs), however the computational requirements for analyzing raw PPG waveforms can be significant. The analysis of PPG-derived peak-to-peak intervals may offer a more feasible solution for smartphone deployment, provided the diagnostic utility is comparable.

AIMS: To compare raw PPG waveforms and PPG-derived peak-to-peak intervals as input signals for machine learning detection of AF.

METHODS: We developed specialized neural networks for raw waveform and peak-to-peak interval analyses and trained them on 7,704 PPGs from 106 patients from the TeleCheck-AF project. We evaluated the neural networks on 48,912 PPGs from 416 patients from the VIRTUAL-SAFARI project. We recorded computational requirements, sensitivity, positive predictive value (PPV), and F1 score.

RESULTS: With 1.6 million trainable parameters, the waveform model was more than 100 times as complex as the interval model (15,513 parameters) and required 19 times more computational power. In external validation, metrics were comparable between the interval and waveform models. For the interval model vs. the waveform model, sensitivity was 91.7 % vs. 81.9 % (p=0.4), PPV was 80.5 % vs. 84.5 % (p=0.3), and F1 score was 85.6 % vs. 81.3 % (p=0.5), respectively.

CONCLUSION: PPG-derived peak-to-peak intervals and PPG waveforms were equivalent as input signals to neural networks in terms of accurate AF detection. The reduced computational requirements of the interval model make it a more suitable option for deployment on digital end-user devices such as smartphones.

PMID:39644781 | DOI:10.1016/j.cmpb.2024.108537

Talanta. 2024 Dec 2;285:127268. doi: 10.1016/j.talanta.2024.127268. Online ahead of print.

ABSTRACT

Foodborne pathogenic infections pose a significant threat to human health. Accurate detection of foodborne diseases is essential in preventing disease transmission. This study proposed an AI model for precisely identifying foodborne pathogenic bacteria based on single-cell Raman spectrum. Self-transfer deep learning and ensemble prediction algorithms had been incorporated into the model framework to improve training efficiency and predictive performance, significantly improving prediction results. Our model can identify simultaneously gram-negative and positive, genus, species of foodborne pathogenic bacteria with an accuracy over 99.99 %, as well as recognized strain with over 99.49 %. At all four classification levels, unprecedented excellent predictive performance had been achieved. This advancement holds practical significance for medical detection and diagnosis of foodborne diseases by reducing false negatives.

PMID:39644671 | DOI:10.1016/j.talanta.2024.127268

Neural Netw. 2024 Dec 4;183:106983. doi: 10.1016/j.neunet.2024.106983. Online ahead of print.

ABSTRACT

Stability analysis is an essential aspect of studying the generalization ability of deep learning, as it involves deriving generalization bounds for stochastic gradient descent-based training algorithms. Adversarial training is the most widely used defense against adversarial attacks. However, previous generalization bounds for adversarial training have not included information regarding data distribution. In this paper, we fill this gap by providing generalization bounds for stochastic gradient descent-based adversarial training that incorporate data distribution information. We utilize the concepts of on-average stability and high-order approximate Lipschitz conditions to examine how changes in data distribution and adversarial budget can affect robust generalization gaps. Our derived generalization bounds for both convex and non-convex losses are at least as good as the uniform stability-based counterparts which do not include data distribution information. Furthermore, our findings demonstrate how distribution shifts from data poisoning attacks can impact robust generalization.

PMID:39644596 | DOI:10.1016/j.neunet.2024.106983

Comput Biol Med. 2024 Dec 6;185:109449. doi: 10.1016/j.compbiomed.2024.109449. Online ahead of print.

ABSTRACT

This survey paper provides a comprehensive analysis of various Deep Learning (DL) techniques and algorithms for detecting protein-protein interactions (PPIs). It examines the scalability, interpretability, accuracy, and efficiency of each technique, offering a detailed empirical and experimental evaluation. Empirically, the techniques are assessed based on four key criteria, while experimentally, they are ranked by specific algorithms and broader methodological categories. Deep Neural Networks (DNNs) demonstrated high accuracy but faced limitations such as overfitting and low interpretability. Convolutional Neural Networks (CNNs) were highly efficient at extracting hierarchical features from biological sequences, while Generative Stochastic Networks (GSNs) excelled in handling uncertainty. Long Short-Term Memory (LSTM) networks effectively captured temporal dependencies within PPI sequences, though they presented scalability challenges. This paper concludes with insights into potential improvements and future directions for advancing DL techniques in PPI identification, highlighting areas where further optimization can enhance performance and applicability.

PMID:39644584 | DOI:10.1016/j.compbiomed.2024.109449

Comput Biol Med. 2024 Dec 6;185:109500. doi: 10.1016/j.compbiomed.2024.109500. Online ahead of print.

ABSTRACT

Emphysema, marked by irreversible lung tissue destruction, poses challenges in progression prediction due to its heterogeneity. Early detection is particularly critical for patients with Alpha-1 Antitrypsin Deficiency (AATD), a genetic disorder reducing ATT protein levels. Heterozygous carriers (PiMS and PiMZ) have variable AAT levels thus complicating their prognosis. This study introduces a novel prognostic model, the Lobe-based Transformer encoder (LobTe), designed to predict the annual change in lung density (ΔALD [g/L-yr]) using CT scans. Utilizing a global self-attention mechanism, LobTe specifically analyzes lobar tissue destruction to forecast disease progression. In parallel, we developed and compared a second model utilizing an LSTM architecture that implements a local subject-specific attention mechanism. Our methodology was validated on a cohort of 2,019 participants from the COPDGene study. The LobTe model demonstrated a small root mean squared error (RMSE=1.73 g/L-yr) and a notable correlation coefficient (ρ=0.61), explaining over 35% of the variability in ΔALD (R2= 0.36). Notably, it achieved a higher correlation coefficient of 0.68 for PiMZ heterozygous carriers, indicating its effectiveness in detecting early emphysema progression among smokers with mild to moderate AAT deficiency. The presented models could serve as a tool for monitoring disease progression and informing treatment strategies in carriers and subjects with AATD. Our code is available at github.com/acil-bwh/LobTe.

PMID:39644582 | DOI:10.1016/j.compbiomed.2024.109500

J Comput Chem. 2025 Jan 5;46(1):e27518. doi: 10.1002/jcc.27518.

ABSTRACT

Predicting drug target binding affinity has huge relevance in Modern drug discovery and drug repositioning processes which assist doctors to come up with new drugs or even use the existing drugs for new target proteins. In silico models, using advanced deep learning techniques could further assist these prediction tasks by providing most prominent drug target pairs. Considering these factors, a deep learning based algorithmic framework is developed in this study to support drug target interaction prediction. The proposed SeqDPI model extract the relevant drug and protein features from the one dimensional Sequential representation of the dataset considered using optimized CNN networks that deploy convolutions on varying length of amino acid subsequence's to capture hidden pattern, the convolved drug- protein features obtained are then used as an input to L2 penalized feed forward neural network which matches the local residue patterns in protein classes with molecular fingerprints of drugs to predict the binding strength for all drug target pairs. The proposed model reduces the convolution strain typically encountered in existing in silico models that utilize complex 3D structures of drug protein datasets. The result shows that the SeqDPI model achieves a mean square error MSE of (0.167) across cross validation folds, outperforming baseline models such as KronRLS (0.406), Simboost (0.226), and DeepPS (0.214). Additionally, SeqDPI attains a high CI score of 0.9114 on the benchmark KIBA dataset, demonstrating its statistical significance and computational efficiency compared to existing methods. This gives the relevance and effectiveness of SeqDPI model in accurately predicting binding affinities while working with simpler one-dimensional data, making it a robust and computationally cost-effective solution for drug-target interaction prediction.

PMID:39644133 | DOI:10.1002/jcc.27518

Environ Monit Assess. 2024 Dec 7;197(1):37. doi: 10.1007/s10661-024-13468-3.

ABSTRACT

In the agricultural sector, sugarcane farming is one of the most organized forms of cultivation. India is the second-largest producer of sugarcane in the world. However, sugarcane crops are highly affected by diseases, which significantly affect crop production. Despite development in deep learning techniques, disease detection remains a challenging and time-consuming task. This paper presents a novel Hybrid Shifted-Vision Transformer approach for the automated classification of sugarcane leaf diseases. The model integrates the Vision Transformer architecture with Hybrid Shifted Windows to effectively capture both local and global features, which is crucial for accurately identifying disease patterns at different spatial scales. To improve feature representation and model performance, self-supervised learning is employed using data augmentation techniques like random rotation, flipping, and occlusion, combined with a jigsaw puzzle task that helps the model learn spatial relationships in images. The method addresses class imbalances in the dataset through stratified sampling, ensuring balanced training and testing sets. The approach is fine-tuned on sugarcane leaf disease datasets using categorical cross-entropy loss, minimizing dissimilarity between predicted probabilities and real labels. Experimental results demonstrate that the Hybrid Shifted-Vision Transformer outperforms traditional models, achieving higher accuracy in disease detection of 98.5%, making it crucial for reliable disease diagnosis and decision-making in agriculture. This architecture enables efficient, large-scale automated sugarcane disease monitoring.

PMID:39643787 | DOI:10.1007/s10661-024-13468-3

Arch Bronconeumol. 2024 Nov 22:S0300-2896(24)00439-3. doi: 10.1016/j.arbres.2024.11.001. Online ahead of print.

ABSTRACT

In this narrative review, we address the ongoing challenges of lung cancer (LC) screening using chest low-dose computerized tomography (LDCT) and explore the contributions of artificial intelligence (AI), in overcoming them. We focus on evaluating the initial (baseline) LDCT examination, which provides a wealth of information relevant to the screening participant's health. This includes the detection of large-size prevalent LC and small-size malignant nodules that are typically diagnosed as LCs upon growth in subsequent annual LDCT scans. Additionally, the baseline LDCT examination provides valuable information about smoking-related comorbidities, including cardiovascular disease, chronic obstructive pulmonary disease, and interstitial lung disease (ILD), by identifying relevant markers. Notably, these comorbidities, despite the slow progression of their markers, collectively exceed LC as ultimate causes of death at follow-up in LC screening participants. Computer-assisted diagnosis tools currently improve the reproducibility of radiologic readings and reduce the false negative rate of LDCT. Deep learning (DL) tools that analyze the radiomic features of lung nodules are being developed to distinguish between benign and malignant nodules. Furthermore, AI tools can predict the risk of LC in the years following a baseline LDCT. AI tools that analyze baseline LDCT examinations can also compute the risk of cardiovascular disease or death, paving the way for personalized screening interventions. Additionally, DL tools are available for assessing osteoporosis and ILD, which helps refine the individual's current and future health profile. The primary obstacles to AI integration into the LDCT screening pathway are the generalizability of performance and the explainability.

PMID:39643515 | DOI:10.1016/j.arbres.2024.11.001