Respir Res. 2024 Apr 18;25(1):167. doi: 10.1186/s12931-024-02793-3.

ABSTRACT

BACKGROUND: Chronic obstructive pulmonary disease (COPD) is a frequently diagnosed yet treatable condition, provided it is identified early and managed effectively. This study aims to develop an advanced COPD diagnostic model by integrating deep learning and radiomics features.

METHODS: We utilized a dataset comprising CT images from 2,983 participants, of which 2,317 participants also provided epidemiological data through questionnaires. Deep learning features were extracted using a Variational Autoencoder, and radiomics features were obtained using the PyRadiomics package. Multi-Layer Perceptrons were used to construct models based on deep learning and radiomics features independently, as well as a fusion model integrating both. Subsequently, epidemiological questionnaire data were incorporated to establish a more comprehensive model. The diagnostic performance of standalone models, the fusion model and the comprehensive model was evaluated and compared using metrics including accuracy, precision, recall, F1-score, Brier score, receiver operating characteristic curves, and area under the curve (AUC).

RESULTS: The fusion model exhibited outstanding performance with an AUC of 0.952, surpassing the standalone models based solely on deep learning features (AUC = 0.844) or radiomics features (AUC = 0.944). Notably, the comprehensive model, incorporating deep learning features, radiomics features, and questionnaire variables demonstrated the highest diagnostic performance among all models, yielding an AUC of 0.971.

CONCLUSION: We developed and implemented a data fusion strategy to construct a state-of-the-art COPD diagnostic model integrating deep learning features, radiomics features, and questionnaire variables. Our data fusion strategy proved effective, and the model can be easily deployed in clinical settings.

TRIAL REGISTRATION: Not applicable. This study is NOT a clinical trial, it does not report the results of a health care intervention on human participants.

PMID:38637823 | DOI:10.1186/s12931-024-02793-3

Sci Rep. 2024 Apr 18;14(1):8996. doi: 10.1038/s41598-024-59578-3.

ABSTRACT

Alzheimer's disease (AD), a neurodegenerative disease that mostly affects the elderly, slowly impairs memory, cognition, and daily tasks. AD has long been one of the most debilitating chronic neurological disorders, affecting mostly people over 65. In this study, we investigated the use of Vision Transformer (ViT) for Magnetic Resonance Image processing in the context of AD diagnosis. ViT was utilized to extract features from MRIs, map them to a feature sequence, perform sequence modeling to maintain interdependencies, and classify features using a time series transformer. The proposed model was evaluated using ADNI T1-weighted MRIs for binary and multiclass classification. Two data collections, Complete 1Yr 1.5T and Complete 3Yr 3T, from the ADNI database were used for training and testing. A random split approach was used, allocating 60% for training and 20% for testing and validation, resulting in sample sizes of (211, 70, 70) and (1378, 458, 458), respectively. The performance of our proposed model was compared to various deep learning models, including CNN with BiL-STM and ViT with Bi-LSTM. The suggested technique diagnoses AD with high accuracy (99.048% for binary and 99.014% for multiclass classification), precision, recall, and F-score. Our proposed method offers researchers an approach to more efficient early clinical diagnosis and interventions.

PMID:38637671 | DOI:10.1038/s41598-024-59578-3

Sci Rep. 2024 Apr 18;14(1):8957. doi: 10.1038/s41598-024-59554-x.

ABSTRACT

A pruned VGG19 model subjected to Axial Coronal Sagittal (ACS) convolutions and a custom VGG16 model are benchmarked to predict 3D fabric descriptors from a set of 2D images. The data used for training and testing are extracted from a set of 600 3D biphase microstructures created numerically. Fabric descriptors calculated from the 3D microstructures constitute the ground truth, while the input data are obtained by slicing the 3D microstructures in each direction of space at regular intervals. The computational cost to train the custom ACS-VGG19 model increases linearly with p (the number of images extracted in each direction of space), and increasing p does not improve the performance of the model - or only does so marginally. The best performing ACS-VGG19 model provides a MAPE of 2 to 5% for the means of aggregate size, aspect ratios and solidity, but cannot be used to estimate orientations. The custom VGG16 yields a MAPE of 2% or less for the means of aggregate size, distance to nearest neighbor, aspect ratios and solidity. The MAPE is less than 3% for the mean roundness, and in the range of 5-7% for the aggregate volume fraction and the mean diagonal components of the orientation matrix. Increasing p improves the performance of the custom VGG16 model, but becomes cost ineffective beyond 3 images per direction. For both models, the aggregate volume fraction is predicted with less accuracy than higher order descriptors, which is attributed to the bias given by the loss function towards highly-correlated descriptors. Both models perform better to predict means than standard deviations, which are noisy quantities. The custom VGG16 model performs better than the pruned version of the ACS-VGG19 model, likely because it contains 3 times (p = 1) to 28 times (p = 10) less parameters than the ACS-VGG19 model, allowing better and faster cnvergence, with less data. The custom VGG16 model predicts the second and third invariants of the orientation matrix with a MAPE of 2.8% and 8.9%, respectively, which suggests that the model can predict orientation descriptors regardless of the orientation of the input images.

PMID:38637651 | DOI:10.1038/s41598-024-59554-x

Sci Rep. 2024 Apr 18;14(1):8924. doi: 10.1038/s41598-024-59735-8.

ABSTRACT

Accurate measurement of abdominal aortic aneurysm is essential for selecting suitable stent-grafts to avoid complications of endovascular aneurysm repair. However, the conventional image-based measurements are inaccurate and time-consuming. We introduce the automated workflow including semantic segmentation with active learning (AL) and measurement using an application programming interface of computer-aided design. 300 patients underwent CT scans, and semantic segmentation for aorta, thrombus, calcification, and vessels was performed in 60-300 cases with AL across five stages using UNETR, SwinUNETR, and nnU-Net consisted of 2D, 3D U-Net, 2D-3D U-Net ensemble, and cascaded 3D U-Net. 7 clinical landmarks were automatically measured for 96 patients. In AL stage 5, 3D U-Net achieved the highest dice similarity coefficient (DSC) with statistically significant differences (p < 0.01) except from the 2D-3D U-Net ensemble and cascade 3D U-Net. SwinUNETR excelled in 95% Hausdorff distance (HD95) with significant differences (p < 0.01) except from UNETR and 3D U-Net. DSC of aorta and calcification were saturated at stage 1 and 4, whereas thrombus and vessels were continuously improved at stage 5. The segmentation time between the manual and AL-corrected segmentation using the best model (3D U-Net) was reduced to 9.51 ± 1.02, 2.09 ± 1.06, 1.07 ± 1.10, and 1.07 ± 0.97 min for the aorta, thrombus, calcification, and vessels, respectively (p < 0.001). All measurement and tortuosity ratio measured - 1.71 ± 6.53 mm and - 0.15 ± 0.25. We developed an automated workflow with semantic segmentation and measurement, demonstrating its efficiency compared to conventional methods.

PMID:38637613 | DOI:10.1038/s41598-024-59735-8

Sci Rep. 2024 Apr 18;14(1):8959. doi: 10.1038/s41598-024-56311-y.

ABSTRACT

In modern wireless communication systems, the scarcity of spectrum resources poses challenges to the performance and efficiency of the system. Spectrum prediction technology can help systems better plan and schedule resources to respond to the dynamic changes in spectrum. Dynamic change in the spectrum refers to the changes in the radio spectrum in a wireless communication system. It means that the available spectrum resources may change at different times and locations. In response to this current situation, this study first constructs a communication collaborative spectrum sensing model using channel aliasing dense connection networks. Then, combining convolutional neural network and gated cyclic unit network in deep learning technology, a communication spectrum prediction model is built. It aims to achieve accurate perception and prediction of spectrum resources through the aforementioned spectrum sensing and prediction models. The results confirm that the proposed perception model has inconsistent perception accuracy under different number of secondary users, with a maximum of 0.99. It is verified that the proposed spectrum prediction model achieves a high prediction accuracy of 0.95 within 208 s and its performance outperforms current similar models. The results are based on the model's deep learning analysis of massive historical communication data, in which the optimized shuffle dense net model plus convolutional gated recurrent unit model is the key to achieve fast and accurate prediction. On the contrary, the highest spectrum prediction accuracy of Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), and Convolutional Neural Networks-Long Short-Term Memory (ConvLSTM) models are 0.86, 0.90, and 0.85, respectively. And the model needs to run for a longer period of time, up to 324, for ConvLSTM to reach the prediction accuracy value of 0.95. In summary, the perception and prediction model built by this research has good performance, and its application in the field of wireless communication can assist staff in better monitoring spectral changes, thereby making more efficient use of spectral resources.

PMID:38637605 | DOI:10.1038/s41598-024-56311-y

Sci Rep. 2024 Apr 19;14(1):9010. doi: 10.1038/s41598-024-58506-9.

ABSTRACT

Tubular injury is the most common cause of acute kidney injury. Histopathological diagnosis may help distinguish between the different types of acute kidney injury and aid in treatment. To date, a limited number of study has used deep-learning models to assist in the histopathological diagnosis of acute kidney injury. This study aimed to perform histopathological segmentation to identify the four structures of acute renal tubular injury using deep-learning models. A segmentation model was used to classify tubule-specific injuries following cisplatin treatment. A total of 45 whole-slide images with 400 generated patches were used in the segmentation model, and 27,478 annotations were created for four classes: glomerulus, healthy tubules, necrotic tubules, and tubules with casts. A segmentation model was developed using the DeepLabV3 architecture with a MobileNetv3-Large backbone to accurately identify the four histopathological structures associated with acute renal tubular injury in PAS-stained mouse samples. In the segmentation model for four structures, the highest Intersection over Union and the Dice coefficient were obtained for the segmentation of the "glomerulus" class, followed by "necrotic tubules," "healthy tubules," and "tubules with cast" classes. The overall performance of the segmentation algorithm for all classes in the test set included an Intersection over Union of 0.7968 and a Dice coefficient of 0.8772. The Dice scores for the glomerulus, healthy tubules, necrotic tubules, and tubules with cast are 91.78 ± 11.09, 87.37 ± 4.02, 88.08 ± 6.83, and 83.64 ± 20.39%, respectively. The utilization of deep learning in a predictive model has demonstrated promising performance in accurately identifying the degree of injured renal tubules. These results may provide new opportunities for the application of the proposed methods to evaluate renal pathology more effectively.

PMID:38637573 | DOI:10.1038/s41598-024-58506-9

Neurosurg Rev. 2024 Apr 19;47(1):170. doi: 10.1007/s10143-024-02366-4.

ABSTRACT

BACKGROUND: Segmentation tools continue to advance, evolving from manual contouring to deep learning. Researchers have utilized segmentation to study a myriad of posterior fossa-related conditions, such as Chiari malformation, trigeminal neuralgia, post-operative pediatric cerebellar mutism syndrome, and Crouzon syndrome. Herein, we present a summary of the current literature on segmentation of the posterior fossa. The review highlights the various segmentation techniques, and their respective strengths and weaknesses, employed along with objectives and outcomes of the various studies reported in the literature.

METHODS: A literature search was conducted in PubMed, Embase, Cochrane, and Web of Science up to November 2023 for articles on segmentation techniques of posterior fossa. The two senior authors searched through databases based on the keywords of the article separately and then enrolled joint articles that met the inclusion and exclusion criteria.

RESULTS: The initial search identified 2205 articles. After applying inclusion and exclusion criteria, 77 articles were selected for full-text review after screening of titles/abstracts. 52 articles were ultimately included in the review. Segmentation techniques included manual, semi-automated, and fully automated (atlas-based, convolutional neural networks). The most common pathology investigated was Chiari malformation.

CONCLUSIONS: Various forms of segmentation techniques have been used to assess posterior fossa volumes/pathologies and each has its advantages and disadvantages. We discuss these nuances and summarize the current state of literature in the context of posterior fossa-associated pathologies.

PMID:38637466 | DOI:10.1007/s10143-024-02366-4

J Imaging Inform Med. 2024 Apr 18. doi: 10.1007/s10278-024-01111-z. Online ahead of print.

ABSTRACT

While dual-energy computed tomography (DECT) technology introduces energy-specific information in clinical practice, single-energy CT (SECT) is predominantly used, limiting the number of people who can benefit from DECT. This study proposed a novel method to generate synthetic low-energy virtual monochromatic images at 50 keV (sVMI50keV) from SECT images using a transformer-based deep learning model, SwinUNETR. Data were obtained from 85 patients who underwent head and neck radiotherapy. Among these, the model was built using data from 70 patients for whom only DECT images were available. The remaining 15 patients, for whom both DECT and SECT images were available, were used to predict from the actual SECT images. We used the SwinUNETR model to generate sVMI50keV. The image quality was evaluated, and the results were compared with those of the convolutional neural network-based model, Unet. The mean absolute errors from the true VMI50keV were 36.5 ± 4.9 and 33.0 ± 4.4 Hounsfield units for Unet and SwinUNETR, respectively. SwinUNETR yielded smaller errors in tissue attenuation values compared with those of Unet. The contrast changes in sVMI50keV generated by SwinUNETR from SECT were closer to those of DECT-derived VMI50keV than the contrast changes in Unet-generated sVMI50keV. This study demonstrated the potential of transformer-based models for generating synthetic low-energy VMIs from SECT images, thereby improving the image quality of head and neck cancer imaging. It provides a practical and feasible solution to obtain low-energy VMIs from SECT data that can benefit a large number of facilities and patients without access to DECT technology.

PMID:38637424 | DOI:10.1007/s10278-024-01111-z

J Imaging Inform Med. 2024 Apr 18. doi: 10.1007/s10278-024-01115-9. Online ahead of print.

ABSTRACT

We aimed to develop and validate a deep convolutional neural network (DCNN) model capable of accurately identifying spondylolysis or spondylolisthesis on lateral or dynamic X-ray images. A total of 2449 lumbar lateral and dynamic X-ray images were collected from two tertiary hospitals. These images were categorized into lumbar spondylolysis (LS), degenerative lumbar spondylolisthesis (DLS), and normal lumbar in a proportional manner. Subsequently, the images were randomly divided into training, validation, and test sets to establish a classification recognition network. The model training and validation process utilized the EfficientNetV2-M network. The model's ability to generalize was assessed by conducting a rigorous evaluation on an entirely independent test set and comparing its performance with the diagnoses made by three orthopedists and three radiologists. The evaluation metrics employed to assess the model's performance included accuracy, sensitivity, specificity, and F1 score. Additionally, the weight distribution of the network was visualized using gradient-weighted class activation mapping (Grad-CAM). For the doctor group, accuracy ranged from 87.9 to 90.0% (mean, 89.0%), precision ranged from 87.2 to 90.5% (mean, 89.0%), sensitivity ranged from 87.1 to 91.0% (mean, 89.2%), specificity ranged from 93.7 to 94.7% (mean, 94.3%), and F1 score ranged from 88.2 to 89.9% (mean, 89.1%). The DCNN model had accuracy of 92.0%, precision of 91.9%, sensitivity of 92.2%, specificity of 95.7%, and F1 score of 92.0%. Grad-CAM exhibited concentrations of highlighted areas in the intervertebral foraminal region. We developed a DCNN model that intelligently distinguished spondylolysis or spondylolisthesis on lumbar lateral or lumbar dynamic radiographs.

PMID:38637423 | DOI:10.1007/s10278-024-01115-9

BJS Open. 2024 Mar 1;8(2):zrae033. doi: 10.1093/bjsopen/zrae033.

ABSTRACT

BACKGROUND: Risk evaluation of lymph node metastasis for early-stage (T1 and T2) colorectal cancers is critical for determining therapeutic strategies. Traditional methods of lymph node metastasis prediction have limited accuracy. This systematic review aimed to review the potential of artificial intelligence in predicting lymph node metastasis in early-stage colorectal cancers.

METHODS: A comprehensive search was performed of papers that evaluated the potential of artificial intelligence in predicting lymph node metastasis in early-stage colorectal cancers. Studies were appraised using the Joanna Briggs Institute tools. The primary outcome was summarizing artificial intelligence models and their accuracy. Secondary outcomes included influential variables and strategies to address challenges.

RESULTS: Of 3190 screened manuscripts, 11 were included, involving 8648 patients from 1996 to 2023. Due to diverse artificial intelligence models and varied metrics, no data synthesis was performed. Models included random forest algorithms, support vector machine, deep learning, artificial neural network, convolutional neural network and least absolute shrinkage and selection operator regression. Artificial intelligence models' area under the curve values ranged from 0.74 to 0.9993 (slide level) and 0.9476 to 0.9956 (single-node level), outperforming traditional clinical guidelines.

CONCLUSION: Artificial intelligence models show promise in predicting lymph node metastasis in early-stage colorectal cancers, potentially refining clinical decisions and improving outcomes.

PROSPERO REGISTRATION NUMBER: CRD42023409094.

PMID:38637299 | DOI:10.1093/bjsopen/zrae033

Neurol Neuroimmunol Neuroinflamm. 2024 May;11(3):e200222. doi: 10.1212/NXI.0000000000200222. Epub 2024 Apr 18.

ABSTRACT

BACKGROUND AND OBJECTIVES: Thalamic atrophy can be used as a proxy for neurodegeneration in multiple sclerosis (MS). Some data point toward thalamic nuclei that could be affected more than others. However, the dynamic of their changes during MS evolution and the mechanisms driving their differential alterations are still uncertain.

METHODS: We paired a large cohort of 1,123 patients with MS with the same number of healthy controls, all scanned with conventional 3D-T1 MRI. To highlight the main atrophic regions at the thalamic nuclei level, we validated a segmentation strategy consisting of deep learning-based synthesis of sequences, which were used for automatic multiatlas segmentation. Then, through a lifespan-based approach, we could model the dynamics of the 4 main thalamic nuclei groups.

RESULTS: All analyses converged toward a higher rate of atrophy for the posterior and medial groups compared with the anterior and lateral groups. We also demonstrated that focal MS white matter lesions were associated with atrophy of groups of nuclei when specifically located within the associated thalamocortical projections. The volumes of the most affected posterior group, but also of the anterior group, were better associated with clinical disability than the volume of the whole thalamus.

DISCUSSION: These findings point toward the thalamic nuclei adjacent to the third ventricle as more susceptible to neurodegeneration during the entire course of MS through potentiation of disconnection effects by regional factors. Because this information can be obtained even from standard T1-weighted MRI, this paves the way toward such an approach for future monitoring of patients with MS.

PMID:38635941 | DOI:10.1212/NXI.0000000000200222

PLoS One. 2024 Apr 18;19(4):e0300142. doi: 10.1371/journal.pone.0300142. eCollection 2024.

ABSTRACT

In view of the strong randomness and non-stationarity of complex system, this study suggests a hybrid multi-strategy prediction technique based on optimized hybrid denoising and deep learning. Firstly, the Sparrow search algorithm (SSA) is used to optimize Variational mode decomposition (VMD) which can decompose the original signal into several Intrinsic mode functions (IMF). Secondly, calculating the Pearson correlation coefficient (PCC) between each IMF component and the original signal, the subsequences with low correlation are eliminated, and the remaining subsequence are denoised by Wavelet soft threshold (WST) method to obtain effective signals. Thirdly, on the basis of the above data noise reduction and reconstruction, our proposal combines Convolutional neural network (CNN) and Bidirectional short-term memory (BiLSTM) model, which is used to analyze the evolution trend of real time sequence data. Finally, we applied the CNN-BiLSTM-SSA-VMD-WST to predict the real time sequence data together with the other methods in order to prove it's effectiveness. The results show that SNR and CC of the SSA-VMD-WST are the largest (the values are 20.2383 and 0.9342). The performance of the CNN-BiLSTM-SSA-VMD-WST are the best, MAE and RMSE are the smallest (which are 0.150 and 0.188), the goodness of fit R2 is the highest(its value is 0.9364). In contrast with other methods, CNN-BiLSTM-SSA-VMD-WST method is more suitable for denoising and prediction of real time series data than the traditional and singular deep learning methods. The proposed method may provide a reliable way for related prediction in various industries.

PMID:38635832 | DOI:10.1371/journal.pone.0300142

PLoS One. 2024 Apr 18;19(4):e0298809. doi: 10.1371/journal.pone.0298809. eCollection 2024.

ABSTRACT

With the rapid development of the Internet, the continuous increase of malware and its variants have brought greatly challenges for cyber security. Due to the imbalance of the data distribution, the research on malware detection focuses on the accuracy of the whole data sample, while ignoring the detection rate of the minority categories' malware. In the dataset sample, the normal data samples account for the majority, while the attacks' malware accounts for the minority. However, the minority categories' attacks will bring great losses to countries, enterprises, or individuals. For solving the problem, this study proposed the GNGS algorithm to construct a new balance dataset for the model algorithm to pay more attention to the feature learning of the minority attacks' malware to improve the detection rate of attacks' malware. The traditional malware detection method is highly dependent on professional knowledge and static analysis, so we used the Self-Attention with Gate mechanism (SAG) based on the Transformer to carry out feature extraction between the local and global features and filter irrelevant noise information, then extracted the long-distance dependency temporal sequence features by the BiGRU network, and obtained the classification results through the SoftMax classifier. In the study, we used the Alibaba Cloud dataset for malware multi-classification. Compared the GSB deep learning network model with other current studies, the experimental results showed that the Gaussian noise generation strategy (GNGS) could solve the unbalanced distribution of minority categories' malware and the SAG-BiGRU algorithm obtained the accuracy rate of 88.7% on the eight-classification, which has better performance than other existing algorithms, and the GSB model also has a good effect on the NSL-KDD dataset, which showed the GSB model is effective for other network intrusion detection.

PMID:38635682 | DOI:10.1371/journal.pone.0298809

PLoS One. 2024 Apr 18;19(4):e0300473. doi: 10.1371/journal.pone.0300473. eCollection 2024.

ABSTRACT

High-resolution imagery and deep learning models have gained increasing importance in land-use mapping. In recent years, several new deep learning network modeling methods have surfaced. However, there has been a lack of a clear understanding of the performance of these models. In this study, we applied four well-established and robust deep learning models (FCN-8s, SegNet, U-Net, and Swin-UNet) to an open benchmark high-resolution remote sensing dataset to compare their performance in land-use mapping. The results indicate that FCN-8s, SegNet, U-Net, and Swin-UNet achieved overall accuracies of 80.73%, 89.86%, 91.90%, and 96.01%, respectively, on the test set. Furthermore, we assessed the generalization ability of these models using two measures: intersection of union and F1 score, which highlight Swin-UNet's superior robustness compared to the other three models. In summary, our study provides a systematic analysis of the classification differences among these four deep learning models through experiments. It serves as a valuable reference for selecting models in future research, particularly in scenarios such as land-use mapping, urban functional area recognition, and natural resource management.

PMID:38635663 | DOI:10.1371/journal.pone.0300473

IEEE J Biomed Health Inform. 2024 Apr 18;PP. doi: 10.1109/JBHI.2024.3390688. Online ahead of print.

ABSTRACT

Congenital heart disease (CHD) is the most frequent birth defect and a leading cause of infant mortality, emphasizing the crucial need for its early diagnosis. Ultrasound is the primary imaging modality for prenatal CHD screening. As a complement to the four-chamber view, the three-vessel view (3VV) plays a vital role in detecting anomalies in the great vessels. However, the interpretation of fetal cardiac ultrasound images is subjective and relies heavily on operator experience, leading to variability in CHD detection rates, particularly in resource-constrained regions. In this study, we propose an automated method for segmenting the pulmonary artery, ascending aorta, and superior vena cava in the 3VV using a novel deep learning network named CoFi-Net. Our network incorporates a coarse-fine collaborative strategy with two parallel branches dedicated to simultaneous global localization and fine segmentation of the vessels. The coarse branch employs a partial decoder to leverage high-level semantic features, enabling global localization of objects and suppression of irrelevant structures. The fine branch utilizes attention-parameterized skip connections to improve feature representations and improve boundary information. The outputs of the two branches are fused to generate accurate vessel segmentations. Extensive experiments conducted on a collected dataset demonstrate the superiority of CoFi-Net compared to state-of-the-art segmentation models for 3VV segmentation, indicating its great potential for enhancing CHD diagnostic efficiency in clinical practice. Furthermore, CoFi-Net outperforms other deep learning models in breast lesion segmentation on a public breast ultrasound dataset, despite not being specifically designed for this task, demonstrating its potential and robustness for various segmentation tasks.

PMID:38635389 | DOI:10.1109/JBHI.2024.3390688

IEEE J Biomed Health Inform. 2024 Apr 18;PP. doi: 10.1109/JBHI.2024.3390804. Online ahead of print.

ABSTRACT

Diffuse large B-cell lymphoma (DLBCL), a cancer of B cells, has been one of the most challenging and complicated diseases because of its considerable variation in clinical behavior, response to therapy, and prognosis. Radiomic features from medical images, such as PET images, have become one of the most valuable features for disease classification or prognosis prediction using learning-based methods. In this paper, a new flexible ensemble deep learning model is proposed for the prognosis prediction of the DLBCL in 18F-FDG PET images. This study proposes the multi-R-signature construction through selected pre-trained deep learning models for predicting progression-free survival (PFS) and overall survival (OS). The proposed method is trained and validated on two datasets from different imaging centers. Through analyzing and comparing the results, the prediction models, including Age, Ann abor stage, Bulky disease, SUVmax, TMTV, and multi-R-signature, achieve the almost best PFS prediction performance (C-index: 0.770, 95% CI: 0.705-0.834, with feature adding fusion method and C-index: 0.764, 95% CI: 0.695-0.832, with feature concatenate fusion method) and OS prediction (C-index: 0.770 (0.692-0.848) and 0.771 (0.694-0.849)) on the validation dataset. The developed multiparametric model could achieve accurate survival risk stratification of DLBCL patients. The outcomes of this study will be helpful for the early identification of high-risk DLBCL patients with refractory relapses and for guiding individualized treatment strategies.

PMID:38635387 | DOI:10.1109/JBHI.2024.3390804

IEEE Trans Med Imaging. 2024 Apr 18;PP. doi: 10.1109/TMI.2024.3390940. Online ahead of print.

ABSTRACT

The lack of reliable biomarkers makes predicting the conversion from intermediate to neovascular age-related macular degeneration (iAMD, nAMD) a challenging task. We develop a Deep Learning (DL) model to predict the future risk of conversion of an eye from iAMD to nAMD from its current OCT scan. Although eye clinics generate vast amounts of longitudinal OCT scans to monitor AMD progression, only a small subset can be manually labeled for supervised DL. To address this issue, we propose Morph-SSL, a novel Self-supervised Learning (SSL) method for longitudinal data. It uses pairs of unlabelled OCT scans from different visits and involves morphing the scan from the previous visit to the next. The Decoder predicts the transformation for morphing and ensures a smooth feature manifold that can generate intermediate scans between visits through linear interpolation. Next, the Morph-SSL trained features are input to a Classifier which is trained in a supervised manner to model the cumulative probability distribution of the time to conversion with a sigmoidal function. Morph-SSL was trained on unlabelled scans of 399 eyes (3570 visits). The Classifier was evaluated with a five-fold cross-validation on 2418 scans from 343 eyes with clinical labels of the conversion date. The Morph-SSL features achieved an AUC of 0.779 in predicting the conversion to nAMD within the next 6 months, outperforming the same network when trained end-to-end from scratch or pre-trained with popular SSL methods. Automated prediction of the future risk of nAMD onset can enable timely treatment and individualized AMD management.

PMID:38635383 | DOI:10.1109/TMI.2024.3390940

Am Nat. 2024 May;203(5):618-627. doi: 10.1086/729422. Epub 2024 Mar 13.

ABSTRACT

AbstractAutonomous sensors provide opportunities to observe organisms across spatial and temporal scales that humans cannot directly observe. By processing large data streams from autonomous sensors with deep learning methods, researchers can make novel and important natural history discoveries. In this study, we combine automated acoustic monitoring with deep learning models to observe breeding-associated activity in the endangered Sierra Nevada yellow-legged frog (Rana sierrae), a behavior that current surveys do not measure. By deploying inexpensive hydrophones and developing a deep learning model to recognize breeding-associated vocalizations, we discover three undocumented R. sierrae vocalization types and find an unexpected temporal pattern of nocturnal breeding-associated vocal activity. This study exemplifies how the combination of autonomous sensor data and deep learning can shed new light on species' natural history, especially during times or in locations where human observation is limited or impossible.

PMID:38635364 | DOI:10.1086/729422

Environ Sci Pollut Res Int. 2024 Apr 18. doi: 10.1007/s11356-024-33233-w. Online ahead of print.

ABSTRACT

Resource recycling is considered necessary for sustainable development, especially in smart cities where increased urbanization and the variety of waste generated require the development of automated waste management models. The development of smart technology offers a possible alternative to traditional waste management techniques that are proving insufficient to reduce the harmful effects of trash on the environment. This paper proposes an intelligent waste classification model to enhance the classification of waste materials, focusing on the critical aspect of waste classification. The proposed model leverages the InceptionV3 deep learning architecture, augmented by multi-objective beluga whale optimization (MBWO) for hyperparameter optimization. In MBWO, sensitivity and specificity evaluation criteria are integrated linearly as the objective function to find the optimal values of the dropout period, learning rate, and batch size. A benchmark dataset, namely TrashNet is adopted to verify the proposed model's performance. By strategically integrating MBWO, the model achieves a considerable increase in accuracy and efficiency in identifying waste materials, contributing to more effective waste management strategies while encouraging sustainable waste management practices. The proposed intelligent waste classification model outperformed the state-of-the-art models with an accuracy of 97.75%, specificity of 99.55%, F1-score of 97.58%, and sensitivity of 98.88%.

PMID:38635097 | DOI:10.1007/s11356-024-33233-w

Lasers Med Sci. 2024 Apr 18;39(1):106. doi: 10.1007/s10103-024-04053-8.

ABSTRACT

The present study proposed a noninvasive, automated, in vivo assessment method based on optical coherence tomography (OCT) and deep learning techniques to qualitatively and quantitatively analyze the biological effects of 2-µm laser-induced skin damage at different irradiation doses. Different doses of 2-µm laser irradiation established a mouse skin damage model, after which the skin-damaged tissues were imaged non-invasively in vivo using OCT. The acquired images were preprocessed to construct the dataset required for deep learning. The deep learning models used were U-Net, DeepLabV3+, PSP-Net, and HR-Net, and the trained models were used to segment the damage images and further quantify the damage volume of mouse skin under different irradiation doses. The comparison of the qualitative and quantitative results of the four network models showed that HR-Net had the best performance, the highest agreement between the segmentation results and real values, and the smallest error in the quantitative assessment of the damage volume. Based on HR-Net to segment the damage image and quantify the damage volume, the irradiation doses 5.41, 9.55, 13.05, 20.85, 32.71, 52.92, 76.71, and 97.24 J/cm² corresponded to a damage volume of 4.58, 12.56, 16.74, 20.88, 24.52, 30.75, 34.13, and 37.32 mm³. The damage volume increased in a radiation dose-dependent manner.

PMID:38634947 | DOI:10.1007/s10103-024-04053-8