Journal Of Digital Imaging[JOURNAL]
Sun
200 papers
Implementing a Streamlined Radiology Workflow to Close the Loop on Incidental Imaging Findings in the Emergency Department.
Fu T, Berlin S, Gupta A … +3 more , Plecha D, Sunshine J, Sommer J
J Digit Imaging · 2023 Jun · PMID 36650302 · Full text
Actionable incidental findings (AIFs) are common imaging findings unrelated to the clinical indication for the imaging test for which follow-up is recommended. Increasing utilization of imaging in the emergency departmen...
Actionable incidental findings (AIFs) are common imaging findings unrelated to the clinical indication for the imaging test for which follow-up is recommended. Increasing utilization of imaging in the emergency department (ED) in recent years has resulted in more patients with AIFs. When these findings are not properly communicated and followed up upon, there is harm to the patient's health outcome as well as possible increased financial costs for the patient, the health system, and potential litigation. Tracking these findings can be difficult, especially so in a large health system. In this report, we detail our experience implementing a closed-loop AIF program within the ED of 11 satellite hospitals of a large academic health system. Our new workflow streamlined radiologist reporting of AIFs through system macros and by using a standardized form integrated into the dictation software. Upon completion of the form, an automatic email is sent to a dedicated nurse navigator who documented the findings and closed the loop by coordinating follow-up imaging or clinic visits with patients, primary care providers, and specialists. Through the new workflow, a total of 1207 incidental finding reports have been submitted from July 2021 to May 2022. The vast majority of AIFs were identified on CT, and the most common categories included lung nodules, pancreas lesions, liver lesions, and other potentially cancerous lesions. At least 10 new cancers have been detected. We hope this report can help guide other health systems in the design of a closed-loop incidental findings program.
Local Axial Scale-Attention for Universal Lesion Detection.
Liu C, Hou Y, Zhao P … +4 more , Guo Z, Li Y, Zhang H, Zhou J
J Digit Imaging · 2023 Jun · PMID 36650301 · Full text
Universal lesion detection (ULD) in computed tomography (CT) images is an important and challenging prerequisite for computer-aided diagnosis (CAD) to find abnormal tissue, such as tumors of lymph nodes, liver tumors, an...
Universal lesion detection (ULD) in computed tomography (CT) images is an important and challenging prerequisite for computer-aided diagnosis (CAD) to find abnormal tissue, such as tumors of lymph nodes, liver tumors, and lymphadenopathy. The key challenge is that lesions have a tiny size and high similarity with non-lesions, which can easily lead to high false positives. Specifically , non-lesions are nearby normal anatomy that include the bowel, vasculature, and mesentery, which decrease the conspicuity of small lesions since they are often hard to differentiate. In this study, we present a novel scale-attention module that enhances feature discrimination between lesion and non-lesion regions by utilizing the domain knowledge of radiologists to reduce false positives effectively. Inspired by the domain knowledge that radiologists tend to divide each CT image into multiple areas, then detect lesions in these smaller areas separately, a local axial scale-attention (LASA) module is proposed to re-weight each pixel in a feature map by aggregating local features from multiple scales adaptively. In addition, to keep the same weight, a combination of axial pixels in the height- and width-axes is designed, attached with position embedding. The model can be used in CNNs easily and flexibly. We test our method on the DeepLesion dataset. The sensitivities at 0.5, 1, 2, 4, 8, and 16 false positives (FPs) per image and average sensitivity at [0.5, 1, 2, 4] are calculated to evaluate the accuracy. The sensitivities are 78.30%, 84.96%, 89.86%, 93.14%, 95.36%, and 95.54% at 0.5, 1, 2, 4, 8, and 16 FPs per image; the average sensitivity is 86.56%, outperforming the former methods. The proposed method enhances feature discrimination between lesion and non-lesion regions by adding LASA modules. These encouraging results illustrate the potential advantage of exploiting the domain knowledge for lesion detection.
Extended Multimodal Flat Detector CT Imaging in Acute Ischemic Stroke: A Pilot Study.
Hoelter P, Lang S, Beuscher V … +4 more , Kallmuenzer B, Manhart M, Schwab S, Doerfler A
J Digit Imaging · 2023 Jun · PMID 36650300 · Full text
By using Flat detector computed tomography (FD-CT), a one-stop-shop approach in the diagnostic workup of acute ischemic stroke (AIS) might be achieved. Although information on upstream vessels is warranted, dedicated FD-...
By using Flat detector computed tomography (FD-CT), a one-stop-shop approach in the diagnostic workup of acute ischemic stroke (AIS) might be achieved. Although information on upstream vessels is warranted, dedicated FD-CT protocols which include the imaging of the cervical vasculature are still lacking. We aimed to prospectively evaluate the implementation of a new multimodal FD-CT protocol including cervical vessel imaging in AIS patients. In total, 16 patients were included in this study. Eight patients with AIS due to large vessel occlusion (LVO) prospectively received a fully multimodal FD-CT imaging, including non-enhanced flat detector computed tomography (NE-FDCT), dynamic perfusion flat detector computed tomography (FD-CTP) and flat detector computed tomography angiography (FD-CTA) including cervical imaging. For comparison of time metrics and image quality, eight AIS patients, which received multimodal CT imaging, were included retrospectively. Although image quality of NE-FDCT and FD-CTA was rated slightly lower than NE-CT and CTA, all FD-CT datasets were of diagnostic quality. Intracerebral hemorrhage exclusion and LVO detection was reliably possible. Median door-to-image time was comparable for the FD-CT group and the control group (CT:30 min, IQR27-58; FD-CT:44.5 min, IQR31-55, p = 0.491). Door-to-groin-puncture time (CT:79.5 min, IQR65-90; FD-CT:59.5 min, IQR51-67; p = 0.016) and image-to-groin-puncture time (CT:44 min, IQR30-50; FD-CT:14 min, IQR12-18; p < 0.001) were significantly shorter, when patients were directly transferred to the angiosuite, where FD-CT took place. Our study indicates that using a new fully multimodal FD-CT approach including imaging of cervical vessels for first-line imaging in AIS patients is feasible and comparable to multimodal CT imaging with substantial potential to streamline the stroke workflow.
Use of Deep Neural Networks in the Detection and Automated Classification of Lesions Using Clinical Images in Ophthalmology, Dermatology, and Oral Medicine-A Systematic Review.
Gomes RFT, Schuch LF, Martins MD … +5 more , Honório EF, de Figueiredo RM, Schmith J, Machado GN, Carrard VC
J Digit Imaging · 2023 Jun · PMID 36650299 · Full text
Artificial neural networks (ANN) are artificial intelligence (AI) techniques used in the automated recognition and classification of pathological changes from clinical images in areas such as ophthalmology, dermatology,...
Artificial neural networks (ANN) are artificial intelligence (AI) techniques used in the automated recognition and classification of pathological changes from clinical images in areas such as ophthalmology, dermatology, and oral medicine. The combination of enterprise imaging and AI is gaining notoriety for its potential benefits in healthcare areas such as cardiology, dermatology, ophthalmology, pathology, physiatry, radiation oncology, radiology, and endoscopic. The present study aimed to analyze, through a systematic literature review, the application of performance of ANN and deep learning in the recognition and automated classification of lesions from clinical images, when comparing to the human performance. The PRISMA 2020 approach (Preferred Reporting Items for Systematic Reviews and Meta-analyses) was used by searching four databases of studies that reference the use of IA to define the diagnosis of lesions in ophthalmology, dermatology, and oral medicine areas. A quantitative and qualitative analyses of the articles that met the inclusion criteria were performed. The search yielded the inclusion of 60 studies. It was found that the interest in the topic has increased, especially in the last 3 years. We observed that the performance of IA models is promising, with high accuracy, sensitivity, and specificity, most of them had outcomes equivalent to human comparators. The reproducibility of the performance of models in real-life practice has been reported as a critical point. Study designs and results have been progressively improved. IA resources have the potential to contribute to several areas of health. In the coming years, it is likely to be incorporated into everyday life, contributing to the precision and reducing the time required by the diagnostic process.
Using Deep Learning to Predict Treatment Response in Patients with Hepatocellular Carcinoma Treated with Y90 Radiation Segmentectomy.
Wagstaff WV, Villalobos A, Gichoya J … +1 more , Kokabi N
J Digit Imaging · 2023 Jun · PMID 36629989 · Full text
Treatment of hepatocellular carcinoma (HCC) with Y90 radioembolization segmentectomy (Y90-RE) demonstrates a tumor dose-response threshold, where dose estimates are highly dependent on accurate SPECT/CT acquisition, regi...
Treatment of hepatocellular carcinoma (HCC) with Y90 radioembolization segmentectomy (Y90-RE) demonstrates a tumor dose-response threshold, where dose estimates are highly dependent on accurate SPECT/CT acquisition, registration, and reconstruction. Any error can result in distorted absorbed dose distributions and inaccurate estimates of treatment success. This study improves upon the voxel-based dosimetry model, one of the most accurate methods available clinically, by using a deep convolutional network ensemble to account for the spatially variable uptake of Y90 within a treated lesion. A retrospective analysis was conducted in patients with HCC who received Y90-RE at a single institution. Seventy-seven patients with 103 lesions met the inclusion criteria: three or fewer tumors, pre- and post treatment MRI, and no prior Y90-RE. Lesions were labeled as complete (n = 57) or incomplete response (n = 46) based on 3-month post treatment MRI and divided by medical record number into a 20% hold-out test set and 80% training set with 5-fold cross-validation. Slice-wise predictions were made from an average ensemble of models and thresholds from the highest accuracy epochs across all five folds. Lesion predictions were made by thresholding all slice predictions through the lesion. When compared to the voxel-based dosimetry model, our model had a higher F1-score (0.72 vs. 0.2), higher accuracy (0.65 vs. 0.60), and higher sensitivity (1.0 vs. 0.11) at predicting complete treatment response. This algorithm has the potential to identify patients with treatment failure who may benefit from earlier follow-up or additional treatment.
Automated Bone Tumor Segmentation and Classification as Benign or Malignant Using Computed Tomographic Imaging.
Yildiz Potter I, Yeritsyan D, Mahar S … +4 more , Wu J, Nazarian A, Vaziri A, Vaziri A
J Digit Imaging · 2023 Jun · PMID 36627518 · Full text
The purpose of this study was to pair computed tomography (CT) imaging and machine learning for automated bone tumor segmentation and classification to aid clinicians in determining the need for biopsy. In this retrospec...
The purpose of this study was to pair computed tomography (CT) imaging and machine learning for automated bone tumor segmentation and classification to aid clinicians in determining the need for biopsy. In this retrospective study (March 2005-October 2020), a dataset of 84 femur CT scans (50 females and 34 males, 20 years and older) with definitive histologic confirmation of bone lesion (71% malignant) were leveraged to perform automated tumor segmentation and classification. Our method involves a deep learning architecture that receives a DICOM slice and predicts (i) a segmentation mask over the estimated tumor region, and (ii) a corresponding class as benign or malignant. Class prediction for each case is then determined via majority voting. Statistical analysis was conducted via fivefold cross validation, with results reported as averages along with 95% confidence intervals. Despite the imbalance between benign and malignant cases in our dataset, our approach attains similar classification performances in specificity (75%) and sensitivity (79%). Average segmentation performance attains 56% Dice score and reaches up to 80% for an image slice in each scan. The proposed approach establishes the first steps in developing an automated deep learning method on bone tumor segmentation and classification from CT imaging. Our approach attains comparable quantitative performance to existing deep learning models using other imaging modalities, including X-ray. Moreover, visual analysis of bone tumor segmentation indicates that our model is capable of learning typical tumor characteristics and provides a promising direction in aiding the clinical decision process for biopsy.
Hybrid Optimization Algorithm Enabled Deep Learning Approach Brain Tumor Segmentation and Classification Using MRI.
Deepa S, Janet J, Sumathi S … +1 more , Ananth JP
J Digit Imaging · 2023 Jun · PMID 36622465 · Full text
The unnatural and uncontrolled increase of brain cells is called brain tumors, leading to human health danger. Magnetic resonance imaging (MRI) is widely applied for classifying and detecting brain tumors, due to its bet...
The unnatural and uncontrolled increase of brain cells is called brain tumors, leading to human health danger. Magnetic resonance imaging (MRI) is widely applied for classifying and detecting brain tumors, due to its better resolution. In general, medical specialists require more details regarding the size, type, and changes in small lesions for effective classification. The timely and exact diagnosis plays a major role in the efficient treatment of patients. Therefore, in this research, an efficient hybrid optimization algorithm is implemented for brain tumor segmentation and classification. The convolutional neural network (CNN) features are extracted to perform a better classification. The classification is performed by considering the extracted features as the input of the deep residual network (DRN), in which the training is performed using the proposed chronological Jaya honey badger algorithm (CJHBA). The proposed CJHBA is the integration of the Jaya algorithm, honey badger algorithm (HBA), and chronological concept. The performance is evaluated using the BRATS 2018 and Figshare datasets, in which the maximum accuracy, sensitivity, and specificity are attained using the BRATS dataset with values 0.9210, 0.9313, and 0.9284, respectively.
Changes in Radiologists' Gaze Patterns Against Lung X-rays with Different Abnormalities: a Randomized Experiment.
Pershin I, Mustafaev T, Ibragimova D … +1 more , Ibragimov B
J Digit Imaging · 2023 Jun · PMID 36622464 · Full text
The workload of some radiologists increased dramatically in the last several, which resulted in a potentially reduced quality of diagnosis. It was demonstrated that diagnostic accuracy of radiologists significantly reduc...
The workload of some radiologists increased dramatically in the last several, which resulted in a potentially reduced quality of diagnosis. It was demonstrated that diagnostic accuracy of radiologists significantly reduces at the end of work shifts. The study aims to investigate how radiologists cover chest X-rays with their gaze in the presence of different chest abnormalities and high workload. We designed a randomized experiment to quantitatively assess how radiologists' image reading patterns change with the radiological workload. Four radiologists read chest X-rays on a radiological workstation equipped with an eye-tracker. The lung fields on the X-rays were automatically segmented with U-Net neural network allowing to measure the lung coverage with radiologists' gaze. The images were randomly split so that each image was shown at a different time to a different radiologist. Regression models were fit to the gaze data to calculate the treads in lung coverage for individual radiologists and chest abnormalities. For the study, a database of 400 chest X-rays with reference diagnoses was assembled. The average lung coverage with gaze ranged from 55 to 65% per radiologist. For every 100 X-rays read, the lung coverage reduced from 1.3 to 7.6% for the different radiologists. The coverage reduction trends were consistent for all abnormalities ranging from 3.4% per 100 X-rays for cardiomegaly to 4.1% per 100 X-rays for atelectasis. The more image radiologists read, the smaller part of the lung fields they cover with the gaze. This pattern is very stable for all abnormality types and is not affected by the exact order the abnormalities are viewed by radiologists. The proposed randomized experiment captured and quantified consistent changes in X-ray reading for different lung abnormalities that occur due to high workload.
Automatic Segmentation of Psoriasis Skin Images Using Adaptive Chimp Optimization Algorithm-Based CNN.
Mohan S, Kasthuri N
J Digit Imaging · 2023 Jun · PMID 36609894 · Full text
Psoriasis is a severe skin disease that is surveyed outwardly by dermatologists. In recent years, computer vision is the major solution for diagnosing the psoriasis skin disease by segmenting the infected skin images. Be...
Psoriasis is a severe skin disease that is surveyed outwardly by dermatologists. In recent years, computer vision is the major solution for diagnosing the psoriasis skin disease by segmenting the infected skin images. Besides, many researchers had presented efficient machine learning techniques for segmenting the psoriasis skin images. Nevertheless, accuracy and time consumption of the model are further to be improved. Thus, in this work, we present adaptive chimp optimization algorithm (AChOA)-based convolutional neural network (CNN) which is introduced for automatic segmentation of psoriasis skin images. After pre-processing, the input images are segmented using AChOA-CNN model where weight and bias values of CNN are optimized with the AChOA. The search ability of ChOA is enhanced by adapting the chaotic sequence based on tent map. At final, from the segmented output images, artifacts are removed by applying the threshold module. From the simulation, we attain 97% of accuracy.
A Comparison of Three Different Deep Learning-Based Models to Predict the MGMT Promoter Methylation Status in Glioblastoma Using Brain MRI.
Faghani S, Khosravi B, Moassefi M … +2 more , Conte GM, Erickson BJ
J Digit Imaging · 2023 Jun · PMID 36604366 · Full text
Glioblastoma (GBM) is the most common primary malignant brain tumor in adults. The standard treatment for GBM consists of surgical resection followed by concurrent chemoradiotherapy and adjuvant temozolomide. O-6-methylg...
Glioblastoma (GBM) is the most common primary malignant brain tumor in adults. The standard treatment for GBM consists of surgical resection followed by concurrent chemoradiotherapy and adjuvant temozolomide. O-6-methylguanine-DNA methyltransferase (MGMT) promoter methylation status is an important prognostic biomarker that predicts the response to temozolomide and guides treatment decisions. At present, the only reliable way to determine MGMT promoter methylation status is through the analysis of tumor tissues. Considering the complications of the tissue-based methods, an imaging-based approach is preferred. This study aimed to compare three different deep learning-based approaches for predicting MGMT promoter methylation status. We obtained 576 T2WI with their corresponding tumor masks, and MGMT promoter methylation status from, The Brain Tumor Segmentation (BraTS) 2021 datasets. We developed three different models: voxel-wise, slice-wise, and whole-brain. For voxel-wise classification, methylated and unmethylated MGMT tumor masks were made into 1 and 2 with 0 background, respectively. We converted each T2WI into 32 × 32 × 32 patches. We trained a 3D-Vnet model for tumor segmentation. After inference, we constructed the whole brain volume based on the patch's coordinates. The final prediction of MGMT methylation status was made by majority voting between the predicted voxel values of the biggest connected component. For slice-wise classification, we trained an object detection model for tumor detection and MGMT methylation status prediction, then for final prediction, we used majority voting. For the whole-brain approach, we trained a 3D Densenet121 for prediction. Whole-brain, slice-wise, and voxel-wise, accuracy was 65.42% (SD 3.97%), 61.37% (SD 1.48%), and 56.84% (SD 4.38%), respectively.
Development of Automatic Portable Pathology Scanner and Its Evaluation for Clinical Practice.
Jiang P, Liu J, Luo Q … +3 more , Pang B, Xiao D, Cao D
J Digit Imaging · 2023 Jun · PMID 36604365 · Full text
Digital pathological scanners transform traditional glass slides into whole slide images (WSIs), which significantly improve the efficiency of pathological diagnosis and promote the development of digital pathology. Howe...
Digital pathological scanners transform traditional glass slides into whole slide images (WSIs), which significantly improve the efficiency of pathological diagnosis and promote the development of digital pathology. However, the huge economic burden limits the spread and application of general WSI scanners in relatively remote and backward regions. In this paper, we develop an automatic portable cytopathology scanner based on mobile internet, Landing-Smart, to avert the above problems. Landing-Smart is a tiny device with a size of 208 mm × 107 mm × 104 mm and a weight of 1.8 kg, which integrates four main components including a smartphone, a glass slide carrier, an electric controller, and an optical imaging unit. By leveraging a simple optical imaging unit to substitute the sophisticated but complex conventional light microscope, the cost of Landing-Smart is less than $3000, much cheaper than general WSI scanners. On the one hand, Landing-Smart utilizes the built-in camera of the smartphone to acquire field of views (FoVs) in the section one by one. On the other hand, it uploads the images to the cloud server in real time via mobile internet, where the image processing and stitching method is implemented to generate the WSI of the cytological sample. The practical assessment of 209 cervical cytological specimens has demonstrated that Landing-Smart is comparable to general digital scanners in cytopathology diagnosis. Landing-Smart provides an effective tool for preliminary cytological screening in underdeveloped areas.
Artificial Intelligence for Detecting Cephalometric Landmarks: A Systematic Review and Meta-analysis.
de Queiroz Tavares Borges Mesquita G, Vieira WA, Vidigal MTC … +5 more , Travençolo BAN, Beaini TL, Spin-Neto R, Paranhos LR, de Brito Júnior RB
J Digit Imaging · 2023 Jun · PMID 36604364 · Full text
Using computer vision through artificial intelligence (AI) is one of the main technological advances in dentistry. However, the existing literature on the practical application of AI for detecting cephalometric landmarks...
Using computer vision through artificial intelligence (AI) is one of the main technological advances in dentistry. However, the existing literature on the practical application of AI for detecting cephalometric landmarks of orthodontic interest in digital images is heterogeneous, and there is no consensus regarding accuracy and precision. Thus, this review evaluated the use of artificial intelligence for detecting cephalometric landmarks in digital imaging examinations and compared it to manual annotation of landmarks. An electronic search was performed in nine databases to find studies that analyzed the detection of cephalometric landmarks in digital imaging examinations with AI and manual landmarking. Two reviewers selected the studies, extracted the data, and assessed the risk of bias using QUADAS-2. Random-effects meta-analyses determined the agreement and precision of AI compared to manual detection at a 95% confidence interval. The electronic search located 7410 studies, of which 40 were included. Only three studies presented a low risk of bias for all domains evaluated. The meta-analysis showed AI agreement rates of 79% (95% CI: 76-82%, I = 99%) and 90% (95% CI: 87-92%, I = 99%) for the thresholds of 2 and 3 mm, respectively, with a mean divergence of 2.05 (95% CI: 1.41-2.69, I = 10%) compared to manual landmarking. The menton cephalometric landmark showed the lowest divergence between both methods (SMD, 1.17; 95% CI, 0.82; 1.53; I = 0%). Based on very low certainty of evidence, the application of AI was promising for automatically detecting cephalometric landmarks, but further studies should focus on testing its strength and validity in different samples.
Identification of Asymptomatic COVID-19 Patients on Chest CT Images Using Transformer-Based or Convolutional Neural Network-Based Deep Learning Models.
Yin M, Liang X, Wang Z … +10 more , Zhou Y, He Y, Xue Y, Gao J, Lin J, Yu C, Liu L, Liu X, Xu C, Zhu J
J Digit Imaging · 2023 Jun · PMID 36596937 · Full text
Novel coronavirus disease 2019 (COVID-19) has rapidly spread throughout the world; however, it is difficult for clinicians to make early diagnoses. This study is to evaluate the feasibility of using deep learning (DL) mo...
Novel coronavirus disease 2019 (COVID-19) has rapidly spread throughout the world; however, it is difficult for clinicians to make early diagnoses. This study is to evaluate the feasibility of using deep learning (DL) models to identify asymptomatic COVID-19 patients based on chest CT images. In this retrospective study, six DL models (Xception, NASNet, ResNet, EfficientNet, ViT, and Swin), based on convolutional neural networks (CNNs) or transformer architectures, were trained to identify asymptomatic patients with COVID-19 on chest CT images. Data from Yangzhou were randomly split into a training set (n = 2140) and an internal-validation set (n = 360). Data from Suzhou was the external-test set (n = 200). Model performance was assessed by the metrics accuracy, recall, and specificity and was compared with the assessments of two radiologists. A total of 2700 chest CT images were collected in this study. In the validation dataset, the Swin model achieved the highest accuracy of 0.994, followed by the EfficientNet model (0.954). The recall and the precision of the Swin model were 0.989 and 1.000, respectively. In the test dataset, the Swin model was still the best and achieved the highest accuracy (0.980). All the DL models performed remarkably better than the two experts. Last, the time on the test set diagnosis spent by two experts-42 min, 17 s (junior); and 29 min, 43 s (senior)-was significantly higher than those of the DL models (all below 2 min). This study evaluated the feasibility of multiple DL models in distinguishing asymptomatic patients with COVID-19 from healthy subjects on chest CT images. It found that a transformer-based model, the Swin model, performed best.
Implementation of Automated Pipeline for Resting-State fMRI Analysis with PACS Integration.
Li XT, Allen JW, Hu R
J Digit Imaging · 2023 Jun · PMID 36596936 · Full text
In recent years, the quantity and complexity of medical imaging acquisition and processing have increased tremendously. The explosion in volume and need for advanced imaging analysis have led to the creation of numerous...
In recent years, the quantity and complexity of medical imaging acquisition and processing have increased tremendously. The explosion in volume and need for advanced imaging analysis have led to the creation of numerous software programs, which have begun to be incorporated into clinical practice for indications such as automated stroke assessment, brain tumor perfusion processing, and hippocampal volume analysis. Despite these advances, there remains a need for specialized, custom-built software for advanced algorithms and new areas of research that is not widely available or adequately integrated in these "out-of-the-box" solutions. The purpose of this paper is to describe the implementation of an image-processing pipeline that is versatile and simple to create, which allows for rapid prototyping of image analysis algorithms and subsequent testing in a clinical environment. This pipeline uses a combination of Orthanc server, custom MATLAB code, and publicly available FMRIB Software Library and RestNeuMap tools to automatically receive and analyze resting-state functional MRI data collected from a custom filter on the MR scanner output. The processed files are then sent directly to Picture Archiving and Communications System (PACS) without the need for user input. This initial experience can serve as a framework for those interested in simple implementation of an automated pipeline customized to clinical needs.
3D-Reconstructed Contact Surface Area and Tumour Volume on Magnetic Resonance Imaging Improve the Prediction of Extraprostatic Extension of Prostate Cancer.
Veerman H, Hoeks CMA, Sluijter JH … +13 more , van der Eijk JA, Boellaard TN, Roeleveld TA, van der Sluis TM, Nieuwenhuijzen JA, Wit E, Rijkhorst EJ, Heymans MW, van Alphen MJA, van Veen RLP, Vis AN, van der Poel HG, van Leeuwen PJ
J Digit Imaging · 2023 Apr · PMID 36547859 · Full text
This study is to determine whether the volume and contact surface area (CSA) of a tumour with an adjacent prostate capsule on MRI in a three-dimensional (3D) model that can predict side-specific extraprostatic extension...
This study is to determine whether the volume and contact surface area (CSA) of a tumour with an adjacent prostate capsule on MRI in a three-dimensional (3D) model that can predict side-specific extraprostatic extension (EPE) at radical prostatectomy (RP). Patients with localised prostate cancer (PCa) who underwent robot-assisted RP between July 2015 and March 2021 were included in this retrospective study. MRI-based 3D prostate models incorporating the PCa volume and location were reconstructed. The tumour volume and surface variables were extracted. For the prostate-to-tumour and tumour-to-prostate CSAs, the areas in which the distances were ≤ 1, ≤ 2, ≤ 3, ≤ 4, and ≤ 5 mm were defined, and their surface (cm) were determined. Differences in prostate sides with and without pathological EPE were analysed. Multivariable logistic regression analysis to find independent predictors of EPE. Overall, 75/302 (25%) prostate sides showed pathological EPE. Prostate sides with EPE had higher cT-stage, higher PSA density, higher percentage of positive biopsy cores, higher biopsy Gleason scores, higher radiological tumour stage, larger tumour volumes, larger prostate CSA, and larger tumour CSA (all p < 0.001). Multivariable logistic regression analysis showed that the radiological tumour stage (p = 0.001), tumour volume (p < 0.001), prostate CSA (p < 0.001), and tumour CSA (p ≤ 0.001) were independent predictors of pathological EPE. A 3D reconstruction of tumour locations in the prostate improves prediction of extraprostatic extension. Tumours with a higher 3D-reconstructed volume, a higher surface area of tumour in contact with the prostate capsule, and higher surface area of prostate capsule in contact with the tumour are at increased risk of side-specific extraprostatic extension.
BirCat Optimization for Automatic Segmentation of Brain Tumors and Pixel Change Detection Using Post-operative MRI Images.
K V S, Sugitha N
J Digit Imaging · 2023 Apr · PMID 36544068 · Full text
There is an emerging need for medical imaging data to provide patients with timely diagnosis. Magnetic resonance imaging (MRI) images based on brain tumor segmentation approaches possess greater importance in planning tr...
There is an emerging need for medical imaging data to provide patients with timely diagnosis. Magnetic resonance imaging (MRI) images based on brain tumor segmentation approaches possess greater importance in planning treatment. Though, mechanizing the process with different imaging conditions and accuracy is a major challenge due to variations in tumor structures. Hence, an efficient optimization-driven classifier, called BirCat optimization-based deep belief network (BirCat-based DBN) is developed to detect brain tumors. The introduced BirCat is devised by incorporating birdswarm algorithm (BSA) into cat swarm optimization (CSO) algorithm and is employed in tuning the DBN classifier. Here, the first step is pre-processing, where noises, as well as artifacts in input image, are eliminated by means of ROI extraction and filtering method. Then, for segmentation, region growing algorithm is used in which the distance is calculated by the modified Bhattacharya measure. Afterward, each segment is adapted for mining the segment-based features and pixel-based features used for classification. Then, the feature vector is formed and given to the DBN classifier, which is tuned with the help of the introduced BirCat for brain tumor detection. The introduced technique effectively determines the regions with the tumor in the input MRI image. Finally, the change detection is evaluated by analyzing the post-operative MRI image and the segmented image by means of pixel mapping strategy with respect to SURF features. The pixel mapping is utilized to evaluate the percentage change in tumor pixels. The proposed BirCat surpassed other prevailing approaches by producing maximal values of specificity, accuracy, sensitivity, F1-score, and Dice score at 0.92, 0.927, 0.938, 0.909, and 0.937, correspondingly, for dataset 2.
A Novel Deep Learning Model Based on Multi-Scale and Multi-View for Detection of Pulmonary Nodules.
Chen Y, Hou X, Yang Y … +3 more , Ge Q, Zhou Y, Nie S
J Digit Imaging · 2023 Apr · PMID 36544067 · Full text
Lung cancer manifests as pulmonary nodules in the early stage. Thus, the early and accurate detection of these nodules is crucial for improving the survival rate of patients. We propose a novel two-stage model for lung n...
Lung cancer manifests as pulmonary nodules in the early stage. Thus, the early and accurate detection of these nodules is crucial for improving the survival rate of patients. We propose a novel two-stage model for lung nodule detection. In the candidate nodule detection stage, a deep learning model based on 3D context information roughly segments the nodules detects the preprocessed image and obtain candidate nodules. In this model, 3D image blocks are input into the constructed model, and it learns the contextual information between the various slices in the 3D image block. The parameters of our model are equivalent to those of a 2D convolutional neural network (CNN), but the model could effectively learn the 3D context information of the nodules. In the false-positive reduction stage, we propose a multi-scale shared convolutional structure model. Our lung detection model has no significant increase in parameters and computation in both stages of multi-scale and multi-view detection. The proposed model was evaluated by using 888 computed tomography (CT) scans from the LIDC-IDRI dataset and achieved a competition performance metric (CPM) score of 0.957. The average detection sensitivity per scan was 0.971/1.0 FP. Furthermore, an average detection sensitivity of 0.933/1.0 FP per scan was achieved based on data from Shanghai Pulmonary Hospital. Our model exhibited a higher detection sensitivity, a lower false-positive rate, and better generalization than current lung nodule detection methods. The method has fewer parameters and less computational complexity, which provides more possibilities for the clinical application of this method.
Deep Learning for Breast MRI Style Transfer with Limited Training Data.
Cao S, Konz N, Duncan J … +1 more , Mazurowski MA
J Digit Imaging · 2023 Apr · PMID 36544066 · Full text
In this work we introduce a novel medical image style transfer method, StyleMapper, that can transfer medical scans to an unseen style with access to limited training data. This is made possible by training our model on...
In this work we introduce a novel medical image style transfer method, StyleMapper, that can transfer medical scans to an unseen style with access to limited training data. This is made possible by training our model on unlimited possibilities of simulated random medical imaging styles on the training set, making our work more computationally efficient when compared with other style transfer methods. Moreover, our method enables arbitrary style transfer: transferring images to styles unseen in training. This is useful for medical imaging, where images are acquired using different protocols and different scanner models, resulting in a variety of styles that data may need to be transferred between. Our model disentangles image content from style and can modify an image's style by simply replacing the style encoding with one extracted from a single image of the target style, with no additional optimization required. This also allows the model to distinguish between different styles of images, including among those that were unseen in training. We propose a formal description of the proposed model. Experimental results on breast magnetic resonance images indicate the effectiveness of our method for style transfer. Our style transfer method allows for the alignment of medical images taken with different scanners into a single unified style dataset, allowing for the training of other downstream tasks on such a dataset for tasks such as classification, object detection and others.
Detecting Distal Radius Fractures Using a Segmentation-Based Deep Learning Model.
Anttila TT, Karjalainen TV, Mäkelä TO … +4 more , Waris EM, Lindfors NC, Leminen MM, Ryhänen JO
J Digit Imaging · 2023 Apr · PMID 36542269 · Full text
Deep learning algorithms can be used to classify medical images. In distal radius fracture treatment, fracture detection and radiographic assessment of fracture displacement are critical steps. The aim of this study was...
Deep learning algorithms can be used to classify medical images. In distal radius fracture treatment, fracture detection and radiographic assessment of fracture displacement are critical steps. The aim of this study was to use pixel-level annotations of fractures to develop a deep learning model for precise distal radius fracture detection. We randomly divided 3785 consecutive emergency wrist radiograph examinations from six hospitals to a training set (3399 examinations) and test set (386 examinations). The training set was used to develop the deep learning model and the test set to assess its validity. The consensus of three hand surgeons was used as the gold standard for the test set. The area under the ROC curve was 0.97 (CI 0.95-0.98) and 0.95 (CI 0.92-0.98) for examinations without a cast. Fractures were identified with higher accuracy in the postero-anterior radiographs than in the lateral radiographs. Our deep learning model performed well in our multi-hospital and multi-radiograph system manufacturer settings. Thus, segmentation-based deep learning models may provide additional benefit. Further research is needed with algorithm comparison and external validation.
BlueLight: An Open Source DICOM Viewer Using Low-Cost Computation Algorithm Implemented with JavaScript Using Advanced Medical Imaging Visualization.
Chen TT, Sun YC, Chu WC … +1 more , Lien CY
J Digit Imaging · 2023 Apr · PMID 36538245 · Full text
Recently, WebGL has been widely used in numerous web-based medical image viewers to present advanced imaging visualization. However, in the scenario of medical imaging, there are many challenges of computation time and m...
Recently, WebGL has been widely used in numerous web-based medical image viewers to present advanced imaging visualization. However, in the scenario of medical imaging, there are many challenges of computation time and memory consumption that limit the use of advanced image renderings, such as volume rendering and multiplanar reformation/reconstruction, in low-cost mobile devices. In this study, we propose a client-side rendering low-cost computation algorithm for common two- and three-dimensional medical imaging visualization implemented by pure JavaScript. Particularly, we used the functions of cascading style sheet transform and combinate with Digital Imaging and Communications in Medicine (DICOM)-related imaging to replace the application programming interface with high computation to reduce the computation time and save memory consumption while launching medical imaging interpretation on web browsers. The results show the proposed algorithm significantly reduced the consumption of central and graphics processing units on various web browsers. The proposed algorithm was implemented in an open-source web-based DICOM viewer BlueLight; the results show that it has sufficient rendering performance to display 3D medical images with DICOM-compliant annotations and has the ability to connect to image archive via DICOMweb as well.Keywords: WebGL, DICOMweb, Multiplanar reconstruction, Volume rendering, DICOM, JavaScript, Zero-footprint.
About
- Frequency
- Sun
- Papers found
- 200
- RSS feed
- Subscribe