Journal Of Digital Imaging[JOURNAL]
Sun
200 papers
A 3D Radiomics-Based Artificial Neural Network Model for Benign Versus Malignant Vertebral Compression Fracture Classification in MRI.
Chiari-Correia NS, Nogueira-Barbosa MH, Chiari-Correia RD … +1 more , Azevedo-Marques PM
J Digit Imaging · 2023 Aug · PMID 37253895 · Full text
To train an artificial neural network model using 3D radiomic features to differentiate benign from malignant vertebral compression fractures (VCFs) on MRI. This retrospective study analyzed sagittal T1-weighted lumbar s...
To train an artificial neural network model using 3D radiomic features to differentiate benign from malignant vertebral compression fractures (VCFs) on MRI. This retrospective study analyzed sagittal T1-weighted lumbar spine MRIs from 91 patients (average age of 64.24 ± 11.75 years) diagnosed with benign or malignant VCFs from 2010 to 2019, of them 47 (51.6%) had benign VCFs and 44 (48.4%) had malignant VCFs. The lumbar fractures were three-dimensionally segmented and had their radiomic features extracted and selected with the wrapper method. The training set consisted of 100 fractured vertebral bodies from 61 patients (average age of 63.2 ± 12.5 years), and the test set was comprised of 30 fractured vertebral bodies from 30 patients (average age of 66.4 ± 9.9 years). Classification was performed with the multilayer perceptron neural network with a back-propagation algorithm. To validate the model, the tenfold cross-validation technique and an independent test set (holdout) were used. The performance of the model was evaluated using the average with a 95% confidence interval for the ROC AUC, accuracy, sensitivity, and specificity (considering the threshold = 0.5). In the internal validation test, the best model reached a ROC AUC of 0.98, an accuracy of 95% (95/100), a sensitivity of 93.5% (43/46), and specificity of 96.3% (52/54). In the validation with independent test set, the model achieved a ROC AUC of 0.97, an accuracy of 93.3% (28/30), a sensitivity of 93.3% (14/15), and a specificity of 93.3% (14/15). The model proposed in this study using radiomic features could differentiate benign from malignant vertebral compression fractures with excellent performance and is promising as an aid to radiologists in the characterization of VCFs.
Global Radiomic Features from Mammography for Predicting Difficult-To-Interpret Normal Cases.
Siviengphanom S, Gandomkar Z, Lewis SJ … +1 more , Brennan PC
J Digit Imaging · 2023 Aug · PMID 37253894 · Full text
This work aimed to investigate whether global radiomic features (GRFs) from mammograms can predict difficult-to-interpret normal cases (NCs). Assessments from 537 readers interpreting 239 normal mammograms were used to c...
This work aimed to investigate whether global radiomic features (GRFs) from mammograms can predict difficult-to-interpret normal cases (NCs). Assessments from 537 readers interpreting 239 normal mammograms were used to categorise cases as 120 difficult-to-interpret and 119 easy-to-interpret based on cases having the highest and lowest difficulty scores, respectively. Using lattice- and squared-based approaches, 34 handcrafted GRFs per image were extracted and normalised. Three classifiers were constructed: (i) CC and (ii) MLO using the GRFs from corresponding craniocaudal and mediolateral oblique images only, based on the random forest technique for distinguishing difficult- from easy-to-interpret NCs, and (iii) CC + MLO using the median predictive scores from both CC and MLO models. Useful GRFs for the CC and MLO models were recognised using a scree test. The CC and MLO models were trained and validated using the leave-one-out-cross-validation. The models' performances were assessed by the AUC and compared using the DeLong test. A Kruskal-Wallis test was used to examine if the 34 GRFs differed between difficult- and easy-to-interpret NCs and if difficulty level based on the traditional breast density (BD) categories differed among 115 low-BD and 124 high-BD NCs. The CC + MLO model achieved higher performance (0.71 AUC) than the individual CC and MLO model alone (0.66 each), but statistically non-significant difference was found (all p > 0.05). Six GRFs were identified to be valuable in describing difficult-to-interpret NCs. Twenty features, when compared between difficult- and easy-to-interpret NCs, differed significantly (p < 0.05). No statistically significant difference was observed in difficulty between low- and high-BD NCs (p = 0.709). GRF mammographic analysis can predict difficult-to-interpret NCs.
A Radiomics Study: Classification of Breast Lesions by Textural Features from Mammography Images.
Letchumanan N, Wong JHD, Tan LK … +4 more , Ab Mumin N, Ng WL, Chan WY, Rahmat K
J Digit Imaging · 2023 Aug · PMID 37253893 · Full text
This study investigates the feasibility of using texture radiomics features extracted from mammography images to distinguish between benign and malignant breast lesions and to classify benign lesions into different categ...
This study investigates the feasibility of using texture radiomics features extracted from mammography images to distinguish between benign and malignant breast lesions and to classify benign lesions into different categories and determine the best machine learning (ML) model to perform the tasks. Six hundred and twenty-two breast lesions from 200 retrospective patient data were segmented and analysed. Three hundred fifty radiomics features were extracted using the Standardized Environment for Radiomics Analysis (SERA) library, one of the radiomics implementations endorsed by the Image Biomarker Standardisation Initiative (IBSI). The radiomics features and selected patient characteristics were used to train selected machine learning models to classify the breast lesions. A fivefold cross-validation was used to evaluate the performance of the ML models and the top 10 most important features were identified. The random forest (RF) ensemble gave the highest accuracy (89.3%) and positive predictive value (66%) and likelihood ratio of 13.5 in categorising benign and malignant lesions. For the classification of benign lesions, the RF model again gave the highest likelihood ratio of 3.4 compared to the other models. Morphological and textural radiomics features were identified as the top 10 most important features from the random forest models. Patient age was also identified as one of the significant features in the RF model. We concluded that machine learning models trained against texture-based radiomics features and patient features give reasonable performance in differentiating benign versus malignant breast lesions. Our study also demonstrated that the radiomics-based machine learning models were able to emulate the visual assessment of mammography lesions, typically used by radiologists, leading to a better understanding of how the machine learning model arrive at their decision.
A Robust and Explainable Structure-Based Algorithm for Detecting the Organ Boundary From Ultrasound Multi-Datasets.
Peng T, Gu Y, Zhang J … +5 more , Dong Y, Di G, Wang W, Zhao J, Cai J
J Digit Imaging · 2023 Aug · PMID 37231289 · Full text
Detecting the organ boundary in an ultrasound image is challenging because of the poor contrast of ultrasound images and the existence of imaging artifacts. In this study, we developed a coarse-to-refinement architecture...
Detecting the organ boundary in an ultrasound image is challenging because of the poor contrast of ultrasound images and the existence of imaging artifacts. In this study, we developed a coarse-to-refinement architecture for multi-organ ultrasound segmentation. First, we integrated the principal curve-based projection stage into an improved neutrosophic mean shift-based algorithm to acquire the data sequence, for which we utilized a limited amount of prior seed point information as the approximate initialization. Second, a distribution-based evolution technique was designed to aid in the identification of a suitable learning network. Then, utilizing the data sequence as the input of the learning network, we achieved the optimal learning network after learning network training. Finally, a scaled exponential linear unit-based interpretable mathematical model of the organ boundary was expressed via the parameters of a fraction-based learning network. The experimental outcomes indicated that our algorithm 1) achieved more satisfactory segmentation outcomes than state-of-the-art algorithms, with a Dice score coefficient value of 96.68 ± 2.2%, a Jaccard index value of 95.65 ± 2.16%, and an accuracy of 96.54 ± 1.82% and 2) discovered missing or blurry areas.
A Lightweight and Robust Framework for Circulating Genetically Abnormal Cells (CACs) Identification Using 4-Color Fluorescence In Situ Hybridization (FISH) Image and Deep Refined Learning.
Xu X, Li C, Lan X … +4 more , Fan X, Lv X, Ye X, Wu T
J Digit Imaging · 2023 Aug · PMID 37231288 · Full text
Circulating genetically abnormal cells (CACs) constitute an important biomarker for cancer diagnosis and prognosis. This biomarker offers high safety, low cost, and high repeatability, which can serve as a key reference...
Circulating genetically abnormal cells (CACs) constitute an important biomarker for cancer diagnosis and prognosis. This biomarker offers high safety, low cost, and high repeatability, which can serve as a key reference in clinical diagnosis. These cells are identified by counting fluorescence signals using 4-color fluorescence in situ hybridization (FISH) technology, which has a high level of stability, sensitivity, and specificity. However, there are some challenges in CACs identification, due to the difference in the morphology and intensity of staining signals. In this concern, we developed a deep learning network (FISH-Net) based on 4-color FISH image for CACs identification. Firstly, a lightweight object detection network based on the statistical information of signal size was designed to improve the clinical detection rate. Secondly, the rotated Gaussian heatmap with a covariance matrix was defined to standardize the staining signals with different morphologies. Then, the heatmap refinement model was proposed to solve the fluorescent noise interference of 4-color FISH image. Finally, an online repetitive training strategy was used to improve the model's feature extraction ability for hard samples (i.e., fracture signal, weak signal, and adjacent signals). The results showed that the precision was superior to 96%, and the sensitivity was higher than 98%, for fluorescent signal detection. Additionally, validation was performed using the clinical samples of 853 patients from 10 centers. The sensitivity was 97.18% (CI 96.72-97.64%) for CACs identification. The number of parameters of FISH-Net was 2.24 M, compared to 36.9 M for the popularly used lightweight network (YOLO-V7s). The detection speed was about 800 times greater than that of a pathologist. In summary, the proposed network was lightweight and robust for CACs identification. It could greatly increase the review accuracy, enhance the efficiency of reviewers, and reduce the review turnaround time during CACs identification.
An Integrated Ensemble Network Model for Skin Abnormality Detection with Combined Textural Features.
Sharafudeen M, S S VC
J Digit Imaging · 2023 Aug · PMID 37231287 · Full text
Melanoma is the most lethal of all skin cancers. This necessitates the need for a machine learning-driven skin cancer detection system to help medical professionals with early detection. We propose an integrated multi-mo...
Melanoma is the most lethal of all skin cancers. This necessitates the need for a machine learning-driven skin cancer detection system to help medical professionals with early detection. We propose an integrated multi-modal ensemble framework that combines deep convolution neural representations with extracted lesion characteristics and patient meta-data. This study intends to integrate transfer-learned image features, global and local textural information, and patient data using a custom generator to diagnose skin cancer accurately. The architecture combines multiple models in a weighted ensemble strategy, which was trained and validated on specific and distinct datasets, namely, HAM10000, BCN20000 + MSK, and the ISIC2020 challenge datasets. They were evaluated on the mean values of precision, recall or sensitivity, specificity, and balanced accuracy metrics. Sensitivity and specificity play a major role in diagnostics. The model achieved sensitivities of 94.15%, 86.69%, and 86.48% and specificity of 99.24%, 97.73%, and 98.51% for each dataset, respectively. Additionally, the accuracy on the malignant classes of the three datasets was 94%, 87.33%, and 89%, which is significantly higher than the physician recognition rate. The results demonstrate that our weighted voting integrated ensemble strategy outperforms existing models and could serve as an initial diagnostic tool for skin cancer.
Improving Accuracy of Pneumonia Classification Using Modified DenseNet.
Wang K, Jiang P, Kong D … +2 more , Sun B, Shen T
J Digit Imaging · 2023 Aug · PMID 37226023 · Full text
Abstract loading — click title to view on PubMed.
Smart IoT in Breast Cancer Detection Using Optimal Deep Learning.
Majji R, G OPP, Rajeswari R … +1 more , R C
J Digit Imaging · 2023 Aug · PMID 37221422 · Full text
IoT in healthcare systems is currently a viable option for providing higher-quality medical care for contemporary e-healthcare. Using an Internet of Things (IoT)-based smart healthcare system, a trustworthy breast cancer...
IoT in healthcare systems is currently a viable option for providing higher-quality medical care for contemporary e-healthcare. Using an Internet of Things (IoT)-based smart healthcare system, a trustworthy breast cancer classification method called Feedback Artificial Crow Search (FACS)-based Shepherd Convolutional Neural Network (ShCNN) is developed in this research. To choose the best routes, the secure routing operation is first carried out using the recommended FACS while taking fitness measures such as distance, energy, link quality, and latency into account. Then, by merging the Crow Search Algorithm (CSA) and Feedback Artificial Tree, the produced FACS is put into practice (FAT). After the completion of routing phase, the breast cancer categorization process is started at the base station. The feature extraction step is then introduced to the pre-processed input mammography image. As a result, it is possible to successfully get features including area, mean, variance, energy, contrast, correlation, skewness, homogeneity, Gray Level Co-occurrence Matrix (GLCM), and Local Gabor Binary Pattern (LGBP). The quality of the image is next enhanced through data augmentation, and finally, the developed FACS algorithm's ShCNN is used to classify breast cancer. The performance of FACS-based ShCNN is examined using six metrics, including energy, delay, accuracy, sensitivity, specificity, and True Positive Rate (TPR), with the maximum energy of 0.562 J, the least delay of 0.452 s, the highest accuracy of 91.56%, the higher sensitivity of 96.10%, the highest specificity of 91.80%, and the maximum TPR of 99.45%.
Views on Augmented Reality, Virtual Reality, and 3D Printing in Modern Medicine and Education: A Qualitative Exploration of Expert Opinion.
Urlings J, de Jong G, Maal T … +1 more , Henssen D
J Digit Imaging · 2023 Aug · PMID 37162654 · Full text
Although an increased usage and development of 3D technologies is observed in healthcare over the last decades, full integration of these technologies remains challenging. The goal of this project is to qualitatively exp...
Although an increased usage and development of 3D technologies is observed in healthcare over the last decades, full integration of these technologies remains challenging. The goal of this project is to qualitatively explore challenges, pearls, and pitfalls of AR/VR/3D printing applications usage in the medical field of a university medical center. Two rounds of face-to-face interviews were conducted using a semi-structured protocol. First an explorative round was held, interviewing medical specialists (8), PhD students (7), 3D technology specialists (5), and university teachers (3). In the second round, twenty employees in high executive functions of relevant departments were interviewed on seven statements that resulted from the first interviewing round. Data analysis was performed using direct content analyses. The first interviewing round resulted in challenges and opportunities in 3D technology usage that were grouped in 5 themes: aims of using AR/VR/3D printing (1), data acquisition (2), data management plans (3), software packages and segmentation tools (4), and output data and reaching end-user (5). The second interviewing round resulted in an overview of ideas and insights on centralization of knowledge, improving implementation of 3D technology in daily healthcare, reimbursement of 3D technologies, recommendations for further studies, and requirement of using certified software. An overview of challenges and opportunities of 3D technologies in healthcare was provided. Well-designed studies on clinical effectiveness, implementation and cost-effectiveness are warranted for further implementation into the clinical setting.
Discrimination Between Glioblastoma and Solitary Brain Metastasis Using Conventional MRI and Diffusion-Weighted Imaging Based on a Deep Learning Algorithm.
Yan Q, Li F, Cui Y … +10 more , Wang Y, Wang X, Jia W, Liu X, Li Y, Chang H, Shi F, Xia Y, Zhou Q, Zeng Q
J Digit Imaging · 2023 Aug · PMID 37156977 · Full text
This study aims to develop and validate a deep learning (DL) model to differentiate glioblastoma from single brain metastasis (BM) using conventional MRI combined with diffusion-weighted imaging (DWI). Preoperative conve...
This study aims to develop and validate a deep learning (DL) model to differentiate glioblastoma from single brain metastasis (BM) using conventional MRI combined with diffusion-weighted imaging (DWI). Preoperative conventional MRI and DWI of 202 patients with solitary brain tumor (104 glioblastoma and 98 BM) were retrospectively obtained between February 2016 and September 2022. The data were divided into training and validation sets in a 7:3 ratio. An additional 32 patients (19 glioblastoma and 13 BM) from a different hospital were considered testing set. Single-MRI-sequence DL models were developed using the 3D residual network-18 architecture in tumoral (T model) and tumoral + peritumoral regions (T&P model). Furthermore, the combination model based on conventional MRI and DWI was developed. The area under the receiver operating characteristic curve (AUC) was used to assess the classification performance. The attention area of the model was visualized as a heatmap by gradient-weighted class activation mapping technique. For the single-MRI-sequence DL model, the T2WI sequence achieved the highest AUC in the validation set with either T models (0.889) or T&P models (0.934). In the combination models of the T&P model, the model of DWI combined with T2WI and contrast-enhanced T1WI showed increased AUC of 0.949 and 0.930 compared with that of single-MRI sequences in the validation set, respectively. And the highest AUC (0.956) was achieved by combined contrast-enhanced T1WI, T2WI, and DWI. In the heatmap, the central region of the tumoral was hotter and received more attention than other areas and was more important for differentiating glioblastoma from BM. A conventional MRI-based DL model could differentiate glioblastoma from solitary BM, and the combination models improved classification performance.
Fast and Frictionless: A Novel Approach to Radiology Appointment Scheduling Using a Mobile App and Recommendation Engine.
Doshi AM, Ostrow D, Gresens A … +5 more , Grimmelmann R, Mazhar S, Neto E, Woodriff M, Recht M
J Digit Imaging · 2023 Aug · PMID 37145249 · Full text
Many outpatient radiology orders are never scheduled, which can result in adverse outcomes. Digital appointment self-scheduling provides convenience, but utilization has been low. The purpose of this study was to develop...
Many outpatient radiology orders are never scheduled, which can result in adverse outcomes. Digital appointment self-scheduling provides convenience, but utilization has been low. The purpose of this study was to develop a "frictionless" scheduling tool and evaluate the impact on utilization. The existing institutional radiology scheduling app was configured to accommodate a frictionless workflow. A recommendation engine used patient residence, past and future appointment data to generate three optimal appointment suggestions. For eligible frictionless orders, recommendations were sent in a text message. Other orders received either a text message for the non-frictionless app scheduling approach or a call-to-schedule text. Scheduling rates by type of text message and scheduling workflow were analyzed. Baseline data for a 3-month period prior to the launch of frictionless scheduling showed that 17% of orders that received an order notification text were scheduled using the app. In an 11-month period after the launch of frictionless scheduling, the rate of app scheduling was greater for orders that received a text message with recommendations (frictionless approach) versus app schedulable orders that received a text without recommendations (29% vs. 14%, p < 0.01). Thirty-nine percent of the orders that received a frictionless text and scheduled using the app used a recommendation. The most common recommendation rules chosen for scheduling included location preference of prior appointments (52%). Among appointments that were scheduled using a day or time preference, 64% were based on a rule using the time of the day. This study showed that frictionless scheduling was associated with an increased rate of app scheduling.
Deep-Stacked Convolutional Neural Networks for Brain Abnormality Classification Based on MRI Images.
Rumala DJ, van Ooijen P, Rachmadi RF … +2 more , Sensusiati AD, Purnama IKE
J Digit Imaging · 2023 Aug · PMID 37145248 · Full text
An automated diagnosis system is crucial for helping radiologists identify brain abnormalities efficiently. The convolutional neural network (CNN) algorithm of deep learning has the advantage of automated feature extract...
An automated diagnosis system is crucial for helping radiologists identify brain abnormalities efficiently. The convolutional neural network (CNN) algorithm of deep learning has the advantage of automated feature extraction beneficial for an automated diagnosis system. However, several challenges in the CNN-based classifiers of medical images, such as a lack of labeled data and class imbalance problems, can significantly hinder the performance. Meanwhile, the expertise of multiple clinicians may be required to achieve accurate diagnoses, which can be reflected in the use of multiple algorithms. In this paper, we present Deep-Stacked CNN, a deep heterogeneous model based on stacked generalization to harness the advantages of different CNN-based classifiers. The model aims to improve robustness in the task of multi-class brain disease classification when we have no opportunity to train single CNNs on sufficient data. We propose two levels of learning processes to obtain the desired model. At the first level, different pre-trained CNNs fine-tuned via transfer learning will be selected as the base classifiers through several procedures. Each base classifier has a unique expert-like character, which provides diversity to the diagnosis outcomes. At the second level, the base classifiers are stacked together through neural network, representing the meta-learner that best combines their outputs and generates the final prediction. The proposed Deep-Stacked CNN obtained an accuracy of 99.14% when evaluated on the untouched dataset. This model shows its superiority over existing methods in the same domain. It also requires fewer parameters and computations while maintaining outstanding performance.
Automatic Spine Segmentation and Parameter Measurement for Radiological Analysis of Whole-Spine Lateral Radiographs Using Deep Learning and Computer Vision.
Kim YT, Jeong TS, Kim YJ … +3 more , Kim WS, Kim KG, Yee GT
J Digit Imaging · 2023 Aug · PMID 37131065 · Full text
Radiographic examination is essential for diagnosing spinal disorders, and the measurement of spino-pelvic parameters provides important information for the diagnosis and treatment planning of spinal sagittal deformities...
Radiographic examination is essential for diagnosing spinal disorders, and the measurement of spino-pelvic parameters provides important information for the diagnosis and treatment planning of spinal sagittal deformities. While manual measurement methods are the golden standard for measuring parameters, they can be time consuming, inefficient, and rater dependent. Previous studies that have used automatic measurement methods to alleviate the downsides of manual measurements showed low accuracy or could not be applied to general films. We propose a pipeline for automated measurement of spinal parameters by combining a Mask R-CNN model for spine segmentation with computer vision algorithms. This pipeline can be incorporated into clinical workflows to provide clinical utility in diagnosis and treatment planning. A total of 1807 lateral radiographs were used for the training (n = 1607) and validation (n = 200) of the spine segmentation model. An additional 200 radiographs, which were also used for validation, were examined by three surgeons to evaluate the performance of the pipeline. Parameters automatically measured by the algorithm in the test set were statistically compared to parameters measured manually by the three surgeons. The Mask R-CNN model achieved an average precision at 50% intersection over union (AP50) of 96.2% and a Dice score of 92.6% for the spine segmentation task in the test set. The mean absolute error values of the spino-pelvic parameters measurement results were within the range of 0.4° (pelvic tilt) to 3.0° (lumbar lordosis, pelvic incidence), and the standard error of estimate was within the range of 0.5° (pelvic tilt) to 4.0° (pelvic incidence). The intraclass correlation coefficient values ranged from 0.86 (sacral slope) to 0.99 (pelvic tilt, sagittal vertical axis).
A Pilot Human Cadaveric Study on Accuracy of the Augmented Reality Surgical Navigation System for Thoracolumbar Pedicle Screw Insertion Using a New Intraoperative Rapid Registration Method.
Cao B, Yuan B, Xu G … +7 more , Zhao Y, Sun Y, Wang Z, Zhou S, Xu Z, Wang Y, Chen X
J Digit Imaging · 2023 Aug · PMID 37131064 · Full text
To evaluate the feasibility and accuracy of AR-assisted pedicle screw placement using a new intraoperative rapid registration method of combining preoperative CT scanning and intraoperative C-arm 2D fluoroscopy in cadave...
To evaluate the feasibility and accuracy of AR-assisted pedicle screw placement using a new intraoperative rapid registration method of combining preoperative CT scanning and intraoperative C-arm 2D fluoroscopy in cadavers. Five cadavers with intact thoracolumbar spines were employed in this study. Intraoperative registration was performed using anteroposterior and lateral views of preoperative CT scanning and intraoperative 2D fluoroscopic images. Patient-specific targeting guides were used for pedicle screw placement from Th1-L5, totaling 166 screws. Instrumentation for each side was randomized (augmented reality surgical navigation (ARSN) vs. C-arm) with an equal distribution of 83 screws in each group. CT was performed to evaluate the accuracy of both techniques by assessing the screw positions and the deviations between the inserted screws and planned trajectories. Postoperative CT showed that 98.80% (82/83) screws in ARSN group and 72.29% (60/83) screws in C-arm group were within the 2-mm safe zone (p < 0.001). The mean time for instrumentation per level in ARSN group was significantly shorter than that in C-arm group (56.17 ± 3.33 s vs. 99.22 ± 9.03 s, p < 0.001). The overall intraoperative registration time was 17.2 ± 3.5 s per segment. AR-based navigation technology can provide surgeons with accurate guidance of pedicle screw insertion and save the operation time by using the intraoperative rapid registration method of combining preoperative CT scanning and intraoperative C-arm 2D fluoroscopy.
Automated Urine Cell Image Classification Model Using Chaotic Mixer Deep Feature Extraction.
Erten M, Tuncer I, Barua PD … +6 more , Yildirim K, Dogan S, Tuncer T, Tan RS, Fujita H, Acharya UR
J Digit Imaging · 2023 Aug · PMID 37131063 · Full text
Microscopic examination of urinary sediments is a common laboratory procedure. Automated image-based classification of urinary sediments can reduce analysis time and costs. Inspired by cryptographic mixing protocols and...
Microscopic examination of urinary sediments is a common laboratory procedure. Automated image-based classification of urinary sediments can reduce analysis time and costs. Inspired by cryptographic mixing protocols and computer vision, we developed an image classification model that combines a novel Arnold Cat Map (ACM)- and fixed-size patch-based mixer algorithm with transfer learning for deep feature extraction. Our study dataset comprised 6,687 urinary sediment images belonging to seven classes: Cast, Crystal, Epithelia, Epithelial nuclei, Erythrocyte, Leukocyte, and Mycete. The developed model consists of four layers: (1) an ACM-based mixer to generate mixed images from resized 224 × 224 input images using fixed-size 16 × 16 patches; (2) DenseNet201 pre-trained on ImageNet1K to extract 1,920 features from each raw input image, and its six corresponding mixed images were concatenated to form a final feature vector of length 13,440; (3) iterative neighborhood component analysis to select the most discriminative feature vector of optimal length 342, determined using a k-nearest neighbor (kNN)-based loss function calculator; and (4) shallow kNN-based classification with ten-fold cross-validation. Our model achieved 98.52% overall accuracy for seven-class classification, outperforming published models for urinary cell and sediment analysis. We demonstrated the feasibility and accuracy of deep feature engineering using an ACM-based mixer algorithm for image preprocessing combined with pre-trained DenseNet201 for feature extraction. The classification model was both demonstrably accurate and computationally lightweight, making it ready for implementation in real-world image-based urine sediment analysis applications.
Degradation Adaption Local-to-Global Transformer for Low-Dose CT Image Denoising.
Wang H, Chi J, Wu C … +2 more , Yu X, Wu H
J Digit Imaging · 2023 Aug · PMID 37118101 · Full text
Computer tomography (CT) has played an essential role in the field of medical diagnosis. Considering the potential risk of exposing patients to X-ray radiations, low-dose CT (LDCT) images have been widely applied in the...
Computer tomography (CT) has played an essential role in the field of medical diagnosis. Considering the potential risk of exposing patients to X-ray radiations, low-dose CT (LDCT) images have been widely applied in the medical imaging field. Since reducing the radiation dose may result in increased noise and artifacts, methods that can eliminate the noise and artifacts in the LDCT image have drawn increasing attentions and produced impressive results over the past decades. However, recent proposed methods mostly suffer from noise remaining, over-smoothing structures, or false lesions derived from noise. To tackle these issues, we propose a novel degradation adaption local-to-global transformer (DALG-Transformer) for restoring the LDCT image. Specifically, the DALG-Transformer is built on self-attention modules which excel at modeling long-range information between image patch sequences. Meanwhile, an unsupervised degradation representation learning scheme is first developed in medical image processing to learn abstract degradation representations of the LDCT images, which can distinguish various degradations in the representation space rather than the pixel space. Then, we introduce a degradation-aware modulated convolution and gated mechanism into the building modules (i.e., multi-head attention and feed-forward network) of each Transformer block, which can bring in the complementary strength of convolution operation to emphasize on the spatially local context. The experimental results show that the DALG-Transformer can provide superior performance in noise removal, structure preservation, and false lesions elimination compared with five existing representative deep networks. The proposed networks may be readily applied to other image processing tasks including image reconstruction, image deblurring, and image super-resolution.
Carimas: An Extensive Medical Imaging Data Processing Tool for Research.
Rainio O, Han C, Teuho J … +7 more , Nesterov SV, Oikonen V, Piirola S, Laitinen T, Tättäläinen M, Knuuti J, Klén R
J Digit Imaging · 2023 Aug · PMID 37106213 · Full text
Carimas is a multi-purpose medical imaging data processing tool, which can be used to visualize, analyze, and model different medical images in research. Originally, it was developed only for positron emission tomography...
Carimas is a multi-purpose medical imaging data processing tool, which can be used to visualize, analyze, and model different medical images in research. Originally, it was developed only for positron emission tomography data in 2009, but the use of this software has extended to many other tomography imaging modalities, such as computed tomography and magnetic resonance imaging. Carimas is especially well-suited for analysis of three- and four-dimensional image data and creating polar maps in modeling of cardiac perfusion. This article explores various parts of Carimas, including its key features, program structure, and application possibilities.
An Improved Convolution Neural Network and Modified Regularized K-Means-Based Automatic Lung Nodule Detection and Classification.
M DL, M DP
J Digit Imaging · 2023 Aug · PMID 37106212 · Full text
If lung cancer is not detected in its initial phases, it can be fatal. However, because of the quantity and structure of its nodules, lung cancer is difficult to detect early. For accurate detections, radiologists requir...
If lung cancer is not detected in its initial phases, it can be fatal. However, because of the quantity and structure of its nodules, lung cancer is difficult to detect early. For accurate detections, radiologists require assistance from automated tools. Numerous expert methods have been created over time to assist radiologists in the diagnosis of lung cancer. However, this requires accurate research. Therefore, in this article, we propose a framework to precisely detect lung cancer by categorizing it between benign and malignant nodules. To achieve this objective, an efficient deep-learning algorithm is presented. The presented technique consists of four stages, namely pre-processing, segmentation, classification, and severity stage analysis. Initially, the collected image is given to the pre-processing stage to eliminate the distortion present in the image. Then, the noise-free image is given to the segmentation stage. For segmentation, in this paper, modified regularized K-means (MRKM) clustering algorithm is presented. After the segmentation process, the segmented nodule image is fed to the classification stage to categorize the nodule as benign or malignant (risk nodule). For classification, an improved convolution neural network (ICNN) is presented. The proposed ICNN is designed by modifying CNN with the integration of the adaptive tree seed optimization (ATSO) algorithm. Finally, the stage identification is carried out based on the size of the nodule and we classify the malignant nodule as S1-S4. The presented technique attained the maximum accuracy of 96.5% and performance compared with existing state-of-art methods.
Automating Angle Measurements on Foot Radiographs in Young Children: Feasibility and Performance of a Convolutional Neural Network Model.
Patton D, Ghosh A, Farkas A … +9 more , Sotardi S, Francavilla M, Venkatakrishna S, Bose S, Ouyang M, Huang H, Davidson R, Sze R, Nguyen J
J Digit Imaging · 2023 Aug · PMID 37099224 · Full text
Measurement of angles on foot radiographs is an important step in the evaluation of malalignment. The objective is to develop a CNN model to measure angles on radiographs, using radiologists' measurements as the referenc...
Measurement of angles on foot radiographs is an important step in the evaluation of malalignment. The objective is to develop a CNN model to measure angles on radiographs, using radiologists' measurements as the reference standard. This IRB-approved retrospective study included 450 radiographs from 216 patients (< 3 years of age). Angles were automatically measured by means of image segmentation followed by angle calculation, according to Simon's approach for measuring pediatric foot angles. A multiclass U-Net model with a ResNet-34 backbone was used for segmentation. Two pediatric radiologists independently measured anteroposterior and lateral talocalcaneal and talo-1 metatarsal angles using the test dataset and recorded the time used for each study. Intraclass correlation coefficients (ICC) were used to compare angle and paired Wilcoxon signed-rank test to compare time between radiologists and the CNN model. There was high spatial overlap between manual and CNN-based automatic segmentations with dice coefficients ranging between 0.81 (lateral 1 metatarsal) and 0.94 (lateral calcaneus). Agreement was higher for angles on the lateral view when compared to the AP view, between radiologists (ICC: 0.93-0.95, 0.85-0.92, respectively) and between radiologists' mean and CNN calculated (ICC: 0.71-0.73, 0.41-0.52, respectively). Automated angle calculation was significantly faster when compared to radiologists' manual measurements (3 ± 2 vs 114 ± 24 s, respectively; P < 0.001). A CNN model can selectively segment immature ossification centers and automatically calculate angles with a high spatial overlap and moderate to substantial agreement when compared to manual methods, and 39 times faster.
Artificial Intelligence Model Trained with Sparse Data to Detect Facial and Cranial Bone Fractures from Head CT.
Wang HC, Wang SC, Yan JL … +1 more , Ko LW
J Digit Imaging · 2023 Aug · PMID 37095310 · Full text
The presence of cranial and facial bone fractures is an important finding on non-enhanced head computed tomography (CT) scans from patients who have sustained head trauma. Some prior studies have proposed automatic crani...
The presence of cranial and facial bone fractures is an important finding on non-enhanced head computed tomography (CT) scans from patients who have sustained head trauma. Some prior studies have proposed automatic cranial fracture detections, but studies on facial fractures are lacking. We propose a deep learning system to automatically detect both cranial and facial bone fractures. Our system incorporated models consisting of YOLOv4 for one-stage fracture detection and improved ResUNet (ResUNet++) for the segmentation of cranial and facial bones. The results from the two models mapped together provided the location of the fracture and the name of the fractured bone as the final output. The training data for the detection model were the soft tissue algorithm images from a total of 1,447 head CT studies (a total of 16,985 images), and the training data for the segmentation model included 1,538 selected head CT images. The trained models were tested on a test dataset consisting of 192 head CT studies (a total of 5,890 images). The overall performance achieved a sensitivity of 88.66%, a precision of 94.51%, and an F1 score of 0.9149. Specifically, the cranial and facial regions were evaluated and resulted in a sensitivity of 84.78% and 80.77%, a precision of 92.86% and 87.50%, and F1 scores of 0.8864 and 0.8400, respectively. The average accuracy for the segmentation labels concerning all predicted fracture bounding boxes was 80.90%. Our deep learning system could accurately detect cranial and facial bone fractures and identify the fractured bone region simultaneously.
About
- Frequency
- Sun
- Papers found
- 200
- RSS feed
- Subscribe