Journal Of Digital Imaging[JOURNAL]
Sun
200 papers
Subject-Specific Automatic Reconstruction of White Matter Tracts.
Meesters S, Landers M, Rutten GJ … +1 more , Florack L
J Digit Imaging · 2023 Dec · PMID 37537513 · Full text
MRI-based tractography is still underexploited and unsuited for routine use in brain tumor surgery due to heterogeneity of methods and functional-anatomical definitions and above all, the lack of a turn-key system. Stand...
MRI-based tractography is still underexploited and unsuited for routine use in brain tumor surgery due to heterogeneity of methods and functional-anatomical definitions and above all, the lack of a turn-key system. Standardization of methods is therefore desirable, whereby an objective and reliable approach is a prerequisite before the results of any automated procedure can subsequently be validated and used in neurosurgical practice. In this work, we evaluated these preliminary but necessary steps in healthy volunteers. Specifically, we evaluated the robustness and reliability (i.e., test-retest reproducibility) of tractography results of six clinically relevant white matter tracts by using healthy volunteer data (N = 136) from the Human Connectome Project consortium. A deep learning convolutional network-based approach was used for individualized segmentation of regions of interest, combined with an evidence-based tractography protocol and appropriate post-tractography filtering. Robustness was evaluated by estimating the consistency of tractography probability maps, i.e., averaged tractograms in normalized space, through the use of a hold-out cross-validation approach. No major outliers were found, indicating a high robustness of the tractography results. Reliability was evaluated at the individual level. First by examining the overlap of tractograms that resulted from repeatedly processed identical MRI scans (N = 10, 10 iterations) to establish an upper limit of reliability of the pipeline. Second, by examining the overlap for subjects that were scanned twice at different time points (N = 40). Both analyses indicated high reliability, with the second analysis showing a reliability near the upper limit. The robust and reliable subject-specific generation of white matter tracts in healthy subjects holds promise for future validation of our pipeline in a clinical population and subsequent implementation in brain tumor surgery.
Reduced Deep Convolutional Activation Features (R-DeCAF) in Histopathology Images to Improve the Classification Performance for Breast Cancer Diagnosis.
Morovati B, Lashgari R, Hajihasani M … +1 more , Shabani H
J Digit Imaging · 2023 Dec · PMID 37532925 · Full text
Breast cancer is the second most common cancer among women worldwide, and the diagnosis by pathologists is a time-consuming procedure and subjective. Computer-aided diagnosis frameworks are utilized to relieve pathologis...
Breast cancer is the second most common cancer among women worldwide, and the diagnosis by pathologists is a time-consuming procedure and subjective. Computer-aided diagnosis frameworks are utilized to relieve pathologist workload by classifying the data automatically, in which deep convolutional neural networks (CNNs) are effective solutions. The features extracted from the activation layer of pre-trained CNNs are called deep convolutional activation features (DeCAF). In this paper, we have analyzed that all DeCAF features are not necessarily led to higher accuracy in the classification task and dimension reduction plays an important role. We have proposed reduced DeCAF (R-DeCAF) for this purpose, and different dimension reduction methods are applied to achieve an effective combination of features by capturing the essence of DeCAF features. This framework uses pre-trained CNNs such as AlexNet, VGG-16, and VGG-19 as feature extractors in transfer learning mode. The DeCAF features are extracted from the first fully connected layer of the mentioned CNNs, and a support vector machine is used for classification. Among linear and nonlinear dimensionality reduction algorithms, linear approaches such as principal component analysis (PCA) represent a better combination among deep features and lead to higher accuracy in the classification task using a small number of features considering a specific amount of cumulative explained variance (CEV) of features. The proposed method is validated using experimental BreakHis and ICIAR datasets. Comprehensive results show improvement in the classification accuracy up to 4.3% with a feature vector size (FVS) of 23 and CEV equal to 0.15.
An Explainable MRI-Radiomic Quantum Neural Network to Differentiate Between Large Brain Metastases and High-Grade Glioma Using Quantum Annealing for Feature Selection.
Felefly T, Roukoz C, Fares G … +9 more , Achkar S, Yazbeck S, Meyer P, Kordahi M, Azoury F, Nasr DN, Nasr E, Noël G, Francis Z
J Digit Imaging · 2023 Dec · PMID 37507581 · Full text
Solitary large brain metastases (LBM) and high-grade gliomas (HGG) are sometimes hard to differentiate on MRI. The management differs significantly between these two entities, and non-invasive methods that help different...
Solitary large brain metastases (LBM) and high-grade gliomas (HGG) are sometimes hard to differentiate on MRI. The management differs significantly between these two entities, and non-invasive methods that help differentiate between them are eagerly needed to avoid potentially morbid biopsies and surgical procedures. We explore herein the performance and interpretability of an MRI-radiomics variational quantum neural network (QNN) using a quantum-annealing mutual-information (MI) feature selection approach. We retrospectively included 423 patients with HGG and LBM (> 2 cm) who had a contrast-enhanced T1-weighted (CE-T1) MRI between 2012 and 2019. After exclusion, 72 HGG and 129 LBM were kept. Tumors were manually segmented, and a 5-mm peri-tumoral ring was created. MRI images were pre-processed, and 1813 radiomic features were extracted. A set of best features based on MI was selected. MI and conditional-MI were embedded into a quadratic unconstrained binary optimization (QUBO) formulation that was mapped to an Ising-model and submitted to D'Wave's quantum annealer to solve for the best combination of 10 features. The 10 selected features were embedded into a 2-qubits QNN using PennyLane library. The model was evaluated for balanced-accuracy (bACC) and area under the receiver operating characteristic curve (ROC-AUC) on the test set. The model performance was benchmarked against two classical models: dense neural networks (DNN) and extreme gradient boosting (XGB). Shapley values were calculated to interpret sample-wise predictions on the test set. The best 10-feature combination included 6 tumor and 4 ring features. For QNN, DNN, and XGB, respectively, training ROC-AUC was 0.86, 0.95, and 0.94; test ROC-AUC was 0.76, 0.75, and 0.79; and test bACC was 0.74, 0.73, and 0.72. The two most influential features were tumor Laplacian-of-Gaussian-GLRLM-Entropy and sphericity. We developed an accurate interpretable QNN model with quantum-informed feature selection to differentiate between LBM and HGG on CE-T1 brain MRI. The model performance is comparable to state-of-the-art classical models.
ExpHBA Deep-IoT: Exponential Honey Badger Optimized Deep Learning For Breast Cancer Detection in IoT Healthcare System.
Rajeswari R, Sriramakrishnan GV, Ch Vidyadhari … +1 more , Kanimozhi KV
J Digit Imaging · 2023 Dec · PMID 37491544 · Full text
Breast cancer (BC) is the most widely found disease among women in the world. The early detection of BC can frequently lessen the mortality rate as well as progress the probability of providing proper treatment. Hence, t...
Breast cancer (BC) is the most widely found disease among women in the world. The early detection of BC can frequently lessen the mortality rate as well as progress the probability of providing proper treatment. Hence, this paper focuses on devising the Exponential Honey Badger Optimization-based Deep Covolutional Neural Network (EHBO-based DCNN) for early identification of BC in the Internet of Things (IoT). Here, the Honey Badger Optimization (HBO) and Exponential Weighted Moving Average (EWMA) algorithms have been combined to create the EHBO. The EHBO is created to transfer the acquired medical data to the base station (BS) by choosing the best cluster heads to categorize the BC. Then, the statistical and texture features are extracted. Further, data augmentation is performed. Finally, the BC classification is done by DCNN. Thus, the observational outcome reveals that the EHBO-based DCNN algorithm attained outstanding performance concerning the testing accuracy, sensitivity, and specificity of 0.9051, 0.8971, and 0.9029, correspondingly. The accuracy of the proposed method is 7.23%, 6.62%, 5.39%, and 3.45% higher than the methods, such as multi-layer perceptron (MLP) classifier, deep learning, support vector machine (SVM), and ensemble-based classifier.
A Novel Classification Model Using Optimal Long Short-Term Memory for Classification of COVID-19 from CT Images.
Vinothini R, Niranjana G, Yakub F
J Digit Imaging · 2023 Dec · PMID 37491543 · Full text
The human respiratory system is affected when an individual is infected with COVID-19, which became a global pandemic in 2020 and affected millions of people worldwide. However, accurate diagnosis of COVID-19 can be chal...
The human respiratory system is affected when an individual is infected with COVID-19, which became a global pandemic in 2020 and affected millions of people worldwide. However, accurate diagnosis of COVID-19 can be challenging due to small variations in typical and COVID-19 pneumonia, as well as the complexities involved in classifying infection regions. Currently, various deep learning (DL)-based methods are being introduced for the automatic detection of COVID-19 using computerized tomography (CT) scan images. In this paper, we propose the pelican optimization algorithm-based long short-term memory (POA-LSTM) method for classifying coronavirus using CT scan images. The data preprocessing technique is used to convert raw image data into a suitable format for subsequent steps. Here, we develop a general framework called no new U-Net (nnU-Net) for region of interest (ROI) segmentation in medical images. We apply a set of heuristic guidelines derived from the domain to systematically optimize the ROI segmentation task, which represents the dataset's key properties. Furthermore, high-resolution net (HRNet) is a standard neural network design developed for feature extraction. HRNet chooses the top-down strategy over the bottom-up method after considering the two options. It first detects the subject, generates a bounding box around the object and then estimates the relevant feature. The POA is used to minimize the subjective influence of manually selected parameters and enhance the LSTM's parameters. Thus, the POA-LSTM is used for the classification process, achieving higher performance for each performance metric such as accuracy, sensitivity, F1-score, precision, and specificity of 99%, 98.67%, 98.88%, 98.72%, and 98.43%, respectively.
ECTransNet: An Automatic Polyp Segmentation Network Based on Multi-scale Edge Complementary.
Liu W, Li Z, Li C … +1 more , Gao H
J Digit Imaging · 2023 Dec · PMID 37491542 · Full text
Colonoscopy is acknowledged as the foremost technique for detecting polyps and facilitating early screening and prevention of colorectal cancer. In clinical settings, the segmentation of polyps from colonoscopy images ho...
Colonoscopy is acknowledged as the foremost technique for detecting polyps and facilitating early screening and prevention of colorectal cancer. In clinical settings, the segmentation of polyps from colonoscopy images holds paramount importance as it furnishes critical diagnostic and surgical information. Nevertheless, the precise segmentation of colon polyp images is still a challenging task owing to the varied sizes and morphological features of colon polyps and the indistinct boundary between polyps and mucosa. In this study, we present a novel network architecture named ECTransNet to address the challenges in polyp segmentation. Specifically, we propose an edge complementary module that effectively fuses the differences between features with multiple resolutions. This enables the network to exchange features across different levels and results in a substantial improvement in the edge fineness of the polyp segmentation. Additionally, we utilize a feature aggregation decoder that leverages residual blocks to adaptively fuse high-order to low-order features. This strategy restores local edges in low-order features while preserving the spatial information of targets in high-order features, ultimately enhancing the segmentation accuracy. According to extensive experiments conducted on ECTransNet, the results demonstrate that this method outperforms most state-of-the-art approaches on five publicly available datasets. Specifically, our method achieved mDice scores of 0.901 and 0.923 on the Kvasir-SEG and CVC-ClinicDB datasets, respectively. On the Endoscene, CVC-ColonDB, and ETIS datasets, we obtained mDice scores of 0.907, 0.766, and 0.728, respectively.
Initial Experience of 10 Imaging Vendors with the IHE SHARAZONE: a New Multivendor Peer-to-Peer Test Service for DICOM Objects.
Nichols S, Laffin B, Parisot C
J Digit Imaging · 2023 Dec · PMID 37488323 · Full text
Alignment of DICOM (Digital Imaging and Communications in Medicine) capabilities among vendors is crucial to improve interoperability in the healthcare industry and advance medical imaging . However, a sustainable model...
Alignment of DICOM (Digital Imaging and Communications in Medicine) capabilities among vendors is crucial to improve interoperability in the healthcare industry and advance medical imaging . However, a sustainable model for sharing DICOM samples is not available. To address this issue, Integrating the Healthcare Enterprise (IHE) has introduced the IHE SHARAZONE, a continuous cross-vendor DICOM data sharing test service. IHE is a highly regarded organization known for profiling standards such as DICOM, HL7 v2 (Health Level Seven, version 2), HL7 CDA (Clinical Document Architecture), and HL7 FHIR (Fast Healthcare Interoperability Resources) into practical solutions for clinical practice. The primary goal of the IHE SHARAZONE is to provide a reliable and consistent cross-vendor DICOM data sharing system. To evaluate its effectiveness, a 5-month pilot was conducted with ten imaging vendors. The pilot concluded with a participant survey, which yielded valuable insights into the initial experience with the IHE SHARAZONE. These findings can inform future improvements and developments to this important service.
Correction to: Lossy Image Compression in a Preclinical Multimodal Imaging Study.
Cunha FF, Blüml V, Zopf LM … +7 more , Walter A, Wagner M, Weninger WJ, Thomaz LA, Tavora LMN, da Silva Cruz LA, Faria SMM
J Digit Imaging · 2023 Oct · PMID 37468697 · Full text
Abstract loading — click title to view on PubMed.
Artificial Intelligence Techniques for Automatic Detection of Peri-implant Marginal Bone Remodeling in Intraoral Radiographs.
Vera M, Gómez-Silva MJ, Vera V … +7 more , López-González CI, Aliaga I, Gascó E, Vera-González V, Pedrera-Canal M, Besada-Portas E, Pajares G
J Digit Imaging · 2023 Oct · PMID 37468696 · Full text
Peri-implantitis can cause marginal bone remodeling around implants. The aim is to develop an automatic image processing approach based on two artificial intelligence (AI) techniques in intraoral (periapical and bitewing...
Peri-implantitis can cause marginal bone remodeling around implants. The aim is to develop an automatic image processing approach based on two artificial intelligence (AI) techniques in intraoral (periapical and bitewing) radiographs to assist dentists in determining bone loss. The first is a deep learning (DL) object-detector (YOLOv3) to roughly identify (no exact localization is required) two objects: prosthesis (crown) and implant (screw). The second is an image understanding-based (IU) process to fine-tune lines on screw edges and to identify significant points (intensity bone changes, intersections between screw and crown). Distances between these points are used to compute bone loss. A total of 2920 radiographs were used for training (50%) and testing (50%) the DL process. The mAP@0.5 metric is used for performance evaluation of DL considering periapical/bitewing and screws/crowns in upper and lower jaws, with scores ranging from 0.537 to 0.898 (sufficient because DL only needs an approximation). The IU performance is assessed with 50% of the testing radiographs through the t test statistical method, obtaining p values of 0.0106 (line fitting) and 0.0213 (significant point detection). The IU performance is satisfactory, as these values are in accordance with the statistical average/standard deviation in pixels for line fitting (2.75/1.01) and for significant point detection (2.63/1.28) according to the expert criteria of dentists, who establish the ground-truth lines and significant points. In conclusion, AI methods have good prospects for automatic bone loss detection in intraoral radiographs to assist dental specialists in diagnosing peri-implantitis.
Residual Deformable Split Channel and Spatial U-Net for Automated Liver and Liver Tumour Segmentation.
Saumiya S, Franklin SW
J Digit Imaging · 2023 Oct · PMID 37464213 · Full text
Accurate segmentation of the liver and liver tumour (LT) is challenging due to its hazy boundaries and large shape variability. Although using U-Net for liver and LT segmentation achieves better results than manual segme...
Accurate segmentation of the liver and liver tumour (LT) is challenging due to its hazy boundaries and large shape variability. Although using U-Net for liver and LT segmentation achieves better results than manual segmentation, it loses spatial and channel features during segmentation, leading to inaccurate liver and LT segmentation. A residual deformable split depth-wise separable U-Net (RDSDSU-Net) is proposed to increase the accuracy of liver and LT segmentation. The residual deformable convolution layer (DCL) with deformable pooling (DP) is used in the encoder as an attention mechanism to adaptively extract liver and LT shape and position characteristics. Afterward, a convolutional spatial and channel features split graph network (CSCFSG-Net) is introduced in the middle processing layer to improve the expression capability of the liver and LT features by capturing spatial and channel features separately and to extract global contextual liver and LT information from spatial and channel features. Sub-pixel convolutions (SPC) are used in the decoder section to prevent the segmentation results from having a chequerboard artefact effect. Also, the residual deformable encoder features are combined with the decoder through summation to avoid increasing the number of feature maps (FM). Finally, the efficiency of the RDSDSU-Net is evaluated on the 3DIRCADb and LiTS datasets. The DICE score of the proposed RDSDSU-Net achieved 98.21% for liver segmentation and 93.25% for LT segmentation on 3DIRCADb. The experimental outcomes illustrate that the proposed RDSDSU-Net model achieved better segmentation results than the existing techniques.
A Domain-Shift Invariant CNN Framework for Cardiac MRI Segmentation Across Unseen Domains.
Patil SS, Ramteke M, Verma M … +4 more , Seth S, Bhargava R, Mittal S, Rathore AS
J Digit Imaging · 2023 Oct · PMID 37430062 · Full text
The emergence of various deep learning approaches in diagnostic medical image segmentation has made machines capable of accomplishing human-level accuracy. However, the generalizability of these architectures across pati...
The emergence of various deep learning approaches in diagnostic medical image segmentation has made machines capable of accomplishing human-level accuracy. However, the generalizability of these architectures across patients from different countries, Magnetic Resonance Imaging (MRI) scans from distinct vendors, and varying imaging conditions remains questionable. In this work, we propose a translatable deep learning framework for diagnostic segmentation of cine MRI scans. This study aims to render the available SOTA (state-of-the-art) architectures domain-shift invariant by utilizing the heterogeneity of multi-sequence cardiac MRI. To develop and test our approach, we curated a diverse group of public datasets and a dataset obtained from private source. We evaluated 3 SOTA CNN (Convolution neural network) architectures i.e., U-Net, Attention-U-Net, and Attention-Res-U-Net. These architectures were first trained on a combination of three different cardiac MRI sequences. Next, we examined the M&M (multi-center & mutli-vendor) challenge dataset to investigate the effect of different training sets on translatability. The U-Net architecture, trained on the multi-sequence dataset, proved to be the most generalizable across multiple datasets during validation on unseen domains. This model attained mean dice scores of 0.81, 0.85, and 0.83 for myocardial wall segmentation after testing on unseen MyoPS (Myocardial Pathology Segmentation) 2020 dataset, AIIMS (All India Institute of Medical Sciences) dataset and M&M dataset, respectively. Our framework achieved Pearson's correlation values of 0.98, 0.99, and 0.95 between the observed and predicted parameters of end diastole volume, end systole volume, and ejection fraction, respectively, on the unseen Indian population dataset.
Virtual Three-Dimensional Model Analysis in the Assessment of the Maxillary and Mandibular Donor Sites on Cone-Beam Computed Tomography Images.
Diarjani SM, Motevasseli S, Dalili Kajan Z
J Digit Imaging · 2023 Oct · PMID 37428281 · Full text
Using the Mimics software to assess the maxillary and mandibular donor sites on cone-beam computed tomography (CBCT) images. This cross-sectional study was conducted on 80 CBCT scans. Data in DICOM format were transferre...
Using the Mimics software to assess the maxillary and mandibular donor sites on cone-beam computed tomography (CBCT) images. This cross-sectional study was conducted on 80 CBCT scans. Data in DICOM format were transferred to the Mimics software version 21, and a maxillary and a mandibular mask according to cortical and cancellous bones were virtually created for each patient based on Hounsfield units (HUs). Three-dimensional models were reconstructed, and boundaries of donor sites, including mandibular symphysis, ramus, coronoid process, zygomatic buttress, and maxillary tuberosity, were defined. Virtual osteotomy was conducted on the 3D models to harvest bone. The volume, thickness, width, and length of harvestable bone from each site were quantified by the software. Data were analyzed by independent t-test, one-way ANOVA, and Tukey's test (alpha = 0.05). The greatest harvestable bone volume and length differences were observed between ramus and tuberosity (P < 0.001). The maximum and minimum harvestable bone volumes were found in symphysis (1753.54 mm) and tuberosity (84.99 mm). The greatest difference in width and thickness was noted between the coronoid process and tuberosity (P < 0.001) and symphysis and buttress (P < 0.001), respectively. Harvestable bone volume from tuberosity, length, width, volume from symphysis, and volume and thickness from the coronoid process was significantly greater in males (P < 0.05). The harvestable bone volume was the highest in symphysis, followed by ramus, coronoid, buttress, and tuberosity. The harvestable bone length and width were the highest in the symphysis and coronoid process, respectively. Maximum harvestable bone thickness was found in symphysis.
Deep Learning-Based Skin Lesion Multi-class Classification with Global Average Pooling Improvement.
Raghavendra PVSP, Charitha C, Begum KG … +1 more , Prasath VBS
J Digit Imaging · 2023 Oct · PMID 37407845 · Full text
Cancerous skin lesions are one of the deadliest diseases that have the ability in spreading across other body parts and organs. Conventionally, visual inspection and biopsy methods are widely used to detect skin cancers....
Cancerous skin lesions are one of the deadliest diseases that have the ability in spreading across other body parts and organs. Conventionally, visual inspection and biopsy methods are widely used to detect skin cancers. However, these methods have some drawbacks, and the prediction is not highly accurate. This is where a dependable automatic recognition system for skin cancers comes into play. With the extensive usage of deep learning in various aspects of medical health, a novel computer-aided dermatologist tool has been suggested for the accurate identification and classification of skin lesions by deploying a novel deep convolutional neural network (DCNN) model that incorporates global average pooling along with preprocessing to discern the skin lesions. The proposed model is trained and tested on the HAM10000 dataset, which contains seven different classes of skin lesions as target classes. The black hat filtering technique has been applied to remove artifacts in the preprocessing stage along with the resampling techniques to balance the data. The performance of the proposed model is evaluated by comparing it with some of the transfer learning models such as ResNet50, VGG-16, MobileNetV2, and DenseNet121. The proposed model provides an accuracy of 97.20%, which is the highest among the previous state-of-art models for multi-class skin lesion classification. The efficacy of the proposed model is also validated by visualizing the results obtained using a graphical user interface (GUI).
Efficient Collaboration Between Radiologists Using the PACS-Integrated Refer Function to Reduce Communication Times.
Lee S, Kim EK, Chung SY … +1 more , Shin HJ
J Digit Imaging · 2023 Oct · PMID 37407844 · Full text
The purpose of this study was to assess the utility of a picture archiving and communication systems (PACS)-integrated refer function for improving collaboration between radiologists and radiographers during daily readin...
The purpose of this study was to assess the utility of a picture archiving and communication systems (PACS)-integrated refer function for improving collaboration between radiologists and radiographers during daily reading sessions. Retrospective analysis was conducted on refers sent by radiologists using a PACS-integrated refer system from March 2020 to December 2021. Refers were categorized according to receiver: radiologists in the same division (intra-division), radiologists in a different division (inter-division), and radiographers. The proportions of answered refers, content of refers, and timing of refer posts were evaluated. Additionally, time intervals in minutes from initial refer post to refer response were assessed to assess the efficiency of the refer system and compared according to receivers using the Mann-Whitney U test. Among a total of 691 refers posted by radiologists, 579 (83.8%) were answered directly using the refer function in PACS. Of the answered refers, 346 refers (59.8%) were made between radiologists, and 173 (50%) were intra-division refers. About the content of refers, about 82.6% of radiologists' refers were about imaging interpretation consultation, and about 98.9% of refers from radiologists to radiographers were for image quality control. The median time interval until refer response was 9 min, and this response time did not differ between intra-division and inter-division refers (p = 0.998). Of the refers that got responses, 74.3% (257/346) were sent among radiologists before official reports were made, and the median time until refer response was 9-10 min. The proportion of refers answered by radiographers was 85.7% (233/272). The median time interval until refer response by radiographers was 87 min for all refers, and 63% were made within 6 h. Therefore, the PACS-integrated refer function can facilitate communication between radiologists for image interpretation and quality control.
Artificial Intelligence Application to Screen Abdominal Aortic Aneurysm Using Computed tomography Angiography.
Spinella G, Fantazzini A, Finotello A … +7 more , Vincenzi E, Boschetti GA, Brutti F, Magliocco M, Pane B, Basso C, Conti M
J Digit Imaging · 2023 Oct · PMID 37407843 · Full text
The aim of our study is to validate a totally automated deep learning (DL)-based segmentation pipeline to screen abdominal aortic aneurysms (AAA) in computed tomography angiography (CTA) scans. We retrospectively evaluat...
The aim of our study is to validate a totally automated deep learning (DL)-based segmentation pipeline to screen abdominal aortic aneurysms (AAA) in computed tomography angiography (CTA) scans. We retrospectively evaluated 73 thoraco-abdominal CTAs (48 AAA and 25 control CTA) by means of a DL-based segmentation pipeline built on a 2.5D convolutional neural network (CNN) architecture to segment lumen and thrombus of the aorta. The maximum aortic diameter of the abdominal tract was compared using a threshold value (30 mm). Blinded manual measurements from a radiologist were done in order to create a true comparison. The screening pipeline was tested on 48 patients with aneurysm and 25 without aneurysm. The average diameter manually measured was 51.1 ± 14.4 mm for patients with aneurysms and 21.7 ± 3.6 mm for patients without aneurysms. The pipeline correctly classified 47 AAA out of 48 and 24 control patients out of 25 with 97% accuracy, 98% sensitivity, and 96% specificity. The automated pipeline of aneurysm measurements in the abdominal tract reported a median error with regard to the maximum abdominal diameter measurement of 1.3 mm. Our approach allowed for the maximum diameter of 51.2 ± 14.3 mm in patients with aneurysm and 22.0 ± 4.0 mm in patients without an aneurysm. The DL-based screening for AAA is a feasible and accurate method, calling for further validation using a larger pool of diagnostic images towards its clinical use.
Automated Rib Fracture Detection on Chest X-Ray Using Contrastive Learning.
Sun H, Wang X, Li Z … +9 more , Liu A, Xu S, Jiang Q, Li Q, Xue Z, Gong J, Chen L, Xiao Y, Liu S
J Digit Imaging · 2023 Oct · PMID 37407842 · Full text
To develop a deep learning-based model for detecting rib fractures on chest X-Ray and to evaluate its performance based on a multicenter study. Chest digital radiography (DR) images from 18,631 subjects were used for the...
To develop a deep learning-based model for detecting rib fractures on chest X-Ray and to evaluate its performance based on a multicenter study. Chest digital radiography (DR) images from 18,631 subjects were used for the training, testing, and validation of the deep learning fracture detection model. We first built a pretrained model, a simple framework for contrastive learning of visual representations (simCLR), using contrastive learning with the training set. Then, simCLR was used as the backbone for a fully convolutional one-stage (FCOS) objective detection network to identify rib fractures from chest X-ray images. The detection performance of the network for four different types of rib fractures was evaluated using the testing set. A total of 127 images from Data-CZ and 109 images from Data-CH with the annotations for four types of rib fractures were used for evaluation. The results showed that for Data-CZ, the sensitivities of the detection model with no pretraining, pretrained ImageNet, and pretrained DR were 0.465, 0.735, and 0.822, respectively, and the average number of false positives per scan was five in all cases. For the Data-CH test set, the sensitivities of three different pretraining methods were 0.403, 0.655, and 0.748. In the identification of four fracture types, the detection model achieved the highest performance for displaced fractures, with sensitivities of 0.873 and 0.774 for the Data-CZ and Data-CH test sets, respectively, with 5 false positives per scan, followed by nondisplaced fractures, buckle fractures, and old fractures. A pretrained model can significantly improve the performance of the deep learning-based rib fracture detection based on X-ray images, which can reduce missed diagnoses and improve the diagnostic efficacy.
Reproducibility of Deep Learning Algorithms Developed for Medical Imaging Analysis: A Systematic Review.
Moassefi M, Rouzrokh P, Conte GM … +13 more , Vahdati S, Fu T, Tahmasebi A, Younis M, Farahani K, Gentili A, Kline T, Kitamura FC, Huo Y, Kuanar S, Younis K, Erickson BJ, Faghani S
J Digit Imaging · 2023 Oct · PMID 37407841 · Full text
Since 2000, there have been more than 8000 publications on radiology artificial intelligence (AI). AI breakthroughs allow complex tasks to be automated and even performed beyond human capabilities. However, the lack of d...
Since 2000, there have been more than 8000 publications on radiology artificial intelligence (AI). AI breakthroughs allow complex tasks to be automated and even performed beyond human capabilities. However, the lack of details on the methods and algorithm code undercuts its scientific value. Many science subfields have recently faced a reproducibility crisis, eroding trust in processes and results, and influencing the rise in retractions of scientific papers. For the same reasons, conducting research in deep learning (DL) also requires reproducibility. Although several valuable manuscript checklists for AI in medical imaging exist, they are not focused specifically on reproducibility. In this study, we conducted a systematic review of recently published papers in the field of DL to evaluate if the description of their methodology could allow the reproducibility of their findings. We focused on the Journal of Digital Imaging (JDI), a specialized journal that publishes papers on AI and medical imaging. We used the keyword "Deep Learning" and collected the articles published between January 2020 and January 2022. We screened all the articles and included the ones which reported the development of a DL tool in medical imaging. We extracted the reported details about the dataset, data handling steps, data splitting, model details, and performance metrics of each included article. We found 148 articles. Eighty were included after screening for articles that reported developing a DL model for medical image analysis. Five studies have made their code publicly available, and 35 studies have utilized publicly available datasets. We provided figures to show the ratio and absolute count of reported items from included studies. According to our cross-sectional study, in JDI publications on DL in medical imaging, authors infrequently report the key elements of their study to make it reproducible.
HCformer: Hybrid CNN-Transformer for LDCT Image Denoising.
Yuan J, Zhou F, Guo Z … +2 more , Li X, Yu H
J Digit Imaging · 2023 Oct · PMID 37386333 · Full text
Low-dose computed tomography (LDCT) is an effective way to reduce radiation exposure for patients. However, it will increase the noise of reconstructed CT images and affect the precision of clinical diagnosis. The majori...
Low-dose computed tomography (LDCT) is an effective way to reduce radiation exposure for patients. However, it will increase the noise of reconstructed CT images and affect the precision of clinical diagnosis. The majority of the current deep learning-based denoising methods are built on convolutional neural networks (CNNs), which concentrate on local information and have little capacity for multiple structures modeling. Transformer structures are capable of computing each pixel's response on a global scale, but their extensive computation requirements prevent them from being widely used in medical image processing. To reduce the impact of LDCT scans on patients, this paper aims to develop an image post-processing method by combining CNN and Transformer structures. This method can obtain a high-quality images from LDCT. A hybrid CNN-Transformer (HCformer) codec network model is proposed for LDCT image denoising. A neighborhood feature enhancement (NEF) module is designed to introduce the local information into the Transformer's operation, and the representation of adjacent pixel information in the LDCT image denoising task is increased. The shifting window method is utilized to lower the computational complexity of the network model and overcome the problems that come with computing the MSA (Multi-head self-attention) process in a fixed window. Meanwhile, W/SW-MSA (Windows/Shifted window Multi-head self-attention) is alternately used in two layers of the Transformer to gain the information interaction between various Transformer layers. This approach can successfully decrease the Transformer's overall computational cost. The AAPM 2016 LDCT grand challenge dataset is employed for ablation and comparison experiments to demonstrate the viability of the proposed LDCT denoising method. Per the experimental findings, HCformer can increase the image quality metrics SSIM, HuRMSE and FSIM from 0.8017, 34.1898, and 0.6885 to 0.8507, 17.7213, and 0.7247, respectively. Additionally, the proposed HCformer algorithm will preserves image details while it reduces noise. In this paper, an HCformer structure is proposed based on deep learning and evaluated by using the AAPM LDCT dataset. Both the qualitative and quantitative comparison results confirm that the proposed HCformer outperforms other methods. The contribution of each component of the HCformer is also confirmed by the ablation experiments. HCformer can combine the advantages of CNN and Transformer, and it has great potential for LDCT image denoising and other tasks.
Functional Connectivity Networks with Latent Distributions for Mild Cognitive Impairment Identification.
Tang Q, Lu Y, Cai B … +1 more , Wang Y
J Digit Imaging · 2023 Oct · PMID 37369942 · Full text
This work presents a novel approach to estimate brain functional connectivity networks via generative learning. Due to the complexity and variability of rs-fMRI signal, we consider it as a random variable, and utilize va...
This work presents a novel approach to estimate brain functional connectivity networks via generative learning. Due to the complexity and variability of rs-fMRI signal, we consider it as a random variable, and utilize variational autoencoder networks to encode it as a confidence distribution in the latent space rather than as a fixed vector, so as to establish the relationship between them. First, the mean time series of each brain region of interest is mapped into a multivariate Gaussian distribution. The correlation between two brain regions is measured by the Jensen-Shannon divergence that describes the statistical similarity between two probability distributions, and then the adjacency matrix is created to indicate the functional connectivity strength of pairwise brain regions. Meanwhile, our findings show that the adjacency matrices obtained at VAE latent spaces of different dimensionalities have good complementarity for MCI identification in precision and recall, and the classification performance can be further boosted by an efficient cascade of classifiers. This proposal constructs brain functional networks from a statistical modeling standpoint, improving the statistical ability of population data and the generalization ability of observation data variability. We evaluate the proposed framework over the task of identifying subjects with MCI from normal controls, and the experimental results on the public dataset show that our method significantly outperforms both the baseline and current state-of-the-art methods.
COVID-19 Severity Prediction from Chest X-ray Images Using an Anatomy-Aware Deep Learning Model.
Nizam NB, Siddiquee SM, Shirin M … +2 more , Bhuiyan MIH, Hasan T
J Digit Imaging · 2023 Oct · PMID 37369941 · Full text
The COVID-19 pandemic has been adversely affecting the patient management systems in hospitals around the world. Radiological imaging, especially chest x-ray and lung Computed Tomography (CT) scans, plays a vital role in...
The COVID-19 pandemic has been adversely affecting the patient management systems in hospitals around the world. Radiological imaging, especially chest x-ray and lung Computed Tomography (CT) scans, plays a vital role in the severity analysis of hospitalized COVID-19 patients. However, with an increasing number of patients and a lack of skilled radiologists, automated assessment of COVID-19 severity using medical image analysis has become increasingly important. Chest x-ray (CXR) imaging plays a significant role in assessing the severity of pneumonia, especially in low-resource hospitals, and is the most frequently used diagnostic imaging in the world. Previous methods that automatically predict the severity of COVID-19 pneumonia mainly focus on feature pooling from pre-trained CXR models without explicitly considering the underlying human anatomical attributes. This paper proposes an anatomy-aware (AA) deep learning model that learns the generic features from x-ray images considering the underlying anatomical information. Utilizing a pre-trained model and lung segmentation masks, the model generates a feature vector including disease-level features and lung involvement scores. We have used four different open-source datasets, along with an in-house annotated test set for training and evaluation of the proposed method. The proposed method improves the geographical extent score by 11% in terms of mean squared error (MSE) while preserving the benchmark result in lung opacity score. The results demonstrate the effectiveness of the proposed AA model in COVID-19 severity prediction from chest X-ray images. The algorithm can be used in low-resource setting hospitals for COVID-19 severity prediction, especially where there is a lack of skilled radiologists.
About
- Frequency
- Sun
- Papers found
- 200
- RSS feed
- Subscribe