Computed Tomography Image Data Sets Research Articles

A Hybrid Deep Learning Model Combining AgresNet, YOLO, and CNN for Lung Tumor Segmentation and Classification

Lung cancer is a malicious tumor that originates in the respiratory tissues when the cells lining the airways of the lungs proliferate and expand uncontrollably. While computed tomography (CT) scans are capable of illuminating problematic areas, they are not proficient at independently diagnosing lung tumors. The task of accurately determining the nodule distribution within the lung CT scans through automatic identification is challenging. As a consequence, the delineation of lung CT images facilitates the identification and classification of lung tumors. This study involved the acquisition of 1097 lung CT image datasets from the Iraq-Oncology Teaching Hospital/National Centre for Cancer Diseases (IQ-OTH/NCCD). In the initial stage of the work, lung CT images are segmented to find malignancies utilizing the U-Net and Attention Gate Residual U-Net (AGRes U-Net) algorithms. The Intersection over Union (IoU) and binary focal loss metrics are applied to evaluate the segmented lung images. AGRes U-Net achieves 97% greater accuracy than the standard U-Net architecture, as determined by performance evaluation. Subsequently, a YOLOv5 network is implemented to annotate the segmented lung CT images with lesions. Thus, the segmented lung CT outputs are labelled for tumors by the YOLOv5 network are fed into the VGG-19 architecture. With an accuracy of 94.8%, this VGG-19 framework distinguishes the lung tumors as normal, benign, and malignant. To compare the performance of the segmented lung output to that of the classified output by means of a convolutional neural network (CNN), the labelled lung CT images are fed as input to a CNN model in the second phase of the work. In conjunction with the Adaptive Moment Estimation (Adam) optimizer, the CNN model is implemented. The Adam optimizer demonstrates a 98% accuracy rate for the classified tumor outputs. The study illustrates that the accuracy of AGResU-Net segmentation for lung cancer detection and classification, at 97%, is nearly equivalent to that of the CNN model, which achieves 98% accuracy.

Engaging grade school learners with an interactive medical imaging activity.

This case report describes a 45-min active learning lesson plan that engages 4th-5th and 6th-8th grade school students in spatial reasoning through a review of medical imaging. The lesson plan reviews different planar orientations and cross-sections of computed tomography (CT) images of familiar objects. The lesson is designed to introduce students to the idea that scientists are key contributors to healthcare, including in medical imaging technologies that facilitate the visualization of internal structures of patients without invasive procedures. The lesson demonstrates the three standard anatomical planes, axial, sagittal and coronal, by guiding students through CT image datasets of various objects. Students then are led in an interactive "dissection" of fruit to compare internal structures with medical images. The lesson plan aligns with key aspects of Next Generation Science Standards and aims to spark interest in the field of medical physics among a young student population through an introduction to imaging technologies. Worksheets and imaging datasets are included as supplementary materials to facilitate interested physicists adapting this work for educational purposes in their own communities, with minimal repeated effort.

Alternative Non-Destructive Approach for Estimating Morphometric Measurements of Chicken Eggs from Tomographic Images with Computer Vision.

The egg has natural barriers that prevent microbiological contamination and promote food safety. The use of non-destructive methods to obtain morphometric measurements of chicken eggs has the potential to replace traditional invasive techniques, offering greater efficiency and accuracy. This paper aims to demonstrate that estimates derived from non-invasive approaches, such as 3D computed tomography (CT) image analysis, can be comparable to conventional destructive methods. To achieve this goal, two widely recognized deep learning architectures, U-Net 3D and Fully Convolutional Networks (FCN) 3D, were modeled to segment and analyze 3D CT images of chicken eggs. A dataset of real CT images was created and labeled, allowing the extraction of important morphometric measurements, including height, width, shell thickness, and volume. The models achieved an accuracy of up to 98.69%, demonstrating their effectiveness compared to results from manual measurements. These findings highlight the potential of CT image analysis, combined with deep learning, as a non-invasive alternative in industrial and research settings. This approach not only minimizes the need for invasive procedures but also offers a scalable and reliable method for egg quality assessment.

Segmentation of Plain CT Image of Ischemic Lesion based on Trans-Swin-UNet

The present study aims to build a hybrid convolutional neural network and transformer UNet-based model, Trans-Swin-UNet, to segment ischemic lesions of the plain computed tomography (CT) image. The model architecture is built based on TransUnet and has four main improvements. First, replace the decoder of TransUNet with a Swin transformer; second, add a Max Attention module into the skip connection; third, design a comprehensive loss function; and last, speed up the segmentation performance. The present study designs two experiments to evaluate the performance of the built model using both the self-collected and public plain CT image datasets. The model optimization experiment evaluates the improvements of Trans-Swin-UNet over TransUnet. The experimental results show that each improvement of the built model can achieve a better performance than TransUNet in terms of dice similarity coefficient (DSC), Jaccard coefficient (JAC), and accuracy (ACC). The comparison experiment compares the built model with four existing UNet-based models. The experimental results show that the built model had a DSC of 0.72±0.01, a JAC of 0.78±0.04, an ACC of 0.75±0.03 using the self-collected plain CT image dataset and a DSC of 0.73±0.02, a JAC of 0.79±0.03, an ACC of 0.76±0.02 using the public plain CT image dataset, achieving the best segmentation performance among five UNet-based neural network models. The two experimental results conclude that the built model could accurately segment ischemic lesions of the plain CT image. The limitations and future work of this study are also discussed.

Efficient pretraining model based on multi-scale local visual field feature reconstruction for PCB CT image element segmentation

Element segmentation is a key step in nondestructive testing of printed circuit boards (PCBs) based on computed tomography (CT) technology. In recent years, there has been rapid development of self-supervised pretraining technology that can obtain general image features without labeled samples and then uses a small amount of labeled samples to solve downstream tasks, which has good potential in PCB element segmentation. At present, a masked image modeling (MIM) pretraining model has been initially applied in PCB CT image element segmentation. However, due to the small and regular size of PCB elements such as vias, wires, and pads, the global visual field has redundancy for a single-element reconstruction, which may damage the performance of the model. Based on this issue, we propose an efficient pretraining model based on multi-scale local visual field feature reconstruction for PCB CT image element segmentation (EMLR-seg). In this model, the teacher-guided MIM pretraining model is introduced into PCB CT image element segmentation for the first time, to our knowledge, and a multi-scale local visual field extraction (MVE) module is proposed to reduce redundancy by focusing on local visual fields. At the same time, a simple four-Transformer-blocks decoder is used. Experiments show that EMLR-seg can achieve 88.6% mIoU on the PCB CT image dataset we proposed, which exceeds 1.2% by the baseline model, and the training time is reduced by 29.6 h, a reduction of 17.4% under the same experimental condition, which reflects the advantage of EMLR-seg in terms of performance and efficiency.

Lung Cancer Detection Using Convolutional Neural Networks

Lung cancer is among the most prevalent and deadly cancers worldwide. Accurate diagnosis and early detection are critical for improving lung cancer patient outcomes and survival rates. Thanks to developments in medical imaging technology, computer-aided diagnosis (CAD) systems have shown a great deal of promise in helping radiologists identify and diagnose lung cancer from medical images. Here, we present the use of convolutional neural networks (CNNs) to create an early detection system (CAD) for lung cancer. The suggested approach uses lung computed tomography (CT) scans as input and use a CNN architecture to extract high-level features from the pictures. We use transfer learning to enhance a CNN model trained on a large dataset of CT images. The CNN model has been taught to determine if a specific CT image contains lung cancer or not. We evaluate the performance of the proposed CAD system on a dataset of CT scans of the lungs from different institutions. The trial's results show that our CNN-based CAD system can reliably and precisely identify lung cancer from CT scans. We also show the comparative performance of our proposed system against the state-of-the-art machine learning methods for lung cancer prediction. In conclusion, the suggested CNN-based deep learning-based CAD system has produced encouraging results for lung cancer detection from CT scans. The approach might help radiologists identify and classify lung cancer early on, leading to better patient outcomes and survival rates. The viability and usefulness of the suggested approach in clinical practice require more study

Computer-aided diagnosis for lung cancer using waterwheel plant algorithm with deep learning

Lung cancer (LC) is a life-threatening and dangerous disease all over the world. However, earlier diagnoses and treatment can save lives. Earlier diagnoses of malevolent cells in the lungs responsible for oxygenating the human body and expelling carbon dioxide due to significant procedures are critical. Even though a computed tomography (CT) scan is the best imaging approach in the healthcare sector, it is challenging for physicians to identify and interpret the tumour from CT scans. LC diagnosis in CT scan using artificial intelligence (AI) can help radiologists in earlier diagnoses, enhance performance, and decrease false negatives. Deep learning (DL) for detecting lymph node contribution on histopathological slides has become popular due to its great significance in patient diagnoses and treatment. This study introduces a computer-aided diagnosis for LC by utilizing the Waterwheel Plant Algorithm with DL (CADLC-WWPADL) approach. The primary aim of the CADLC-WWPADL approach is to classify and identify the existence of LC on CT scans. The CADLC-WWPADL method uses a lightweight MobileNet model for feature extraction. Besides, the CADLC-WWPADL method employs WWPA for the hyperparameter tuning process. Furthermore, the symmetrical autoencoder (SAE) model is utilized for classification. An investigational evaluation is performed to demonstrate the significant detection outputs of the CADLC-WWPADL technique. An extensive comparative study reported that the CADLC-WWPADL technique effectively performs with other models with a maximum accuracy of 99.05% under the benchmark CT image dataset.

Kidney Tumor Segmentation Based on DWR-SegFormer

Kidney cancer is a malignant tumor with a high mortality rate. The accurate segmentation of tumors from computed tomography (CT) scans can assist physicians in clinical diagnosis. We introduced a new segmentation network called DWR-SegFormer to address the challenge of accurately segmenting kidney tumors in CT images. The method involved binarizing the label maps of clear cell renal cell carcinoma and papillary renal cell carcinoma CT images for identification, and the cancer lesion area was obtained by the label so that the model could accurately identify the area and enhance the feature extraction ability. Secondly, an optimized segmentation model combining a DWR attention mechanism and SegFormer network was constructed. MiT-B0 was used as the encoder of the model to establish long-distance feature dependencies and effectively extract feature information at different resolutions. The decoder with a multi-branch DWR module was implemented to utilize multi-scale feature information effectively and enhance segmentation accuracy. Comparing the experimental results with other existing models shows that the model significantly outperformed the comparison methods in all evaluation metrics on the CT image dataset of clear cell renal cancer. Furthermore, the experimental findings highlight the robustness of the proposed model across other datasets.

Development and Validation of a Diagnosis System for Lung Infection Using Hybrid Deep-Learning Techniques

A fast and accurate test is necessary to detect COVID-19. A computed tomography (CT) scan has shown diagnostic accuracy. CT scan processing using a deep learning architecture may improve illness diagnosis and treatment. We proposed a deep learning system for COVID-19 detection using CT images, including using and comparing transfer-learning, fine-tuning, and the embedding process. This paper presents the development of a COVID-19 case identification model using deep learning techniques. The suggested model utilized a modified visual geometry group (VGG16) architecture as the deep learning framework. The model was trained and validated using a chest CT image dataset. The SARS-COV-2 dataset contains 2482 CT scans of 210 patients from publicly available sources. The modified model demonstrated encouraging outcomes by greatly enhancing the sensitivity measure (95.82±1.75)%, which is an essential criterion for accurately detecting instances of COVID-19 infection. In addition, the model achieved higher values for the accuracy metric (91.67±1.68)%, the specificity meter (88.08±3.72)%, the precision metric (87.51±3.27)%, the F1 score (91.43±1.55)%, and the area under the curve (91.98±1.55)%. Deep learning effectively detects COVID-19 in chest CT scan images. Clinical practitioners may employ the suggested approach to study, identify, and effectively mitigate a greater number of pandemics.

Puncture Accuracy of Robot-Assisted CT-Based Punctures in Interventional Radiology: An Ex Vivo Study.

The purpose of this study was to assess the performance of an optically tracked robot for computed-tomography (CT)-guided needle placements in a phantom study. In total, 240 needle punctures were carried out with the help of an optically tracked robotic device (Micromate) based on CT image datasets at three different slice thicknesses (1, 3, and 5 mm). Conically shaped targets inside a gelatin-filled plexiglass phantom were punctured. The target positioning error between the planned and actual needle trajectory was assessed by measuring the lateral positioning error (ND) between the target and the puncture needle and the Euclidean distance (ED) between the needle tip and target in control CTs. The mean ND and ED for the thinnest CT slice thickness were 1.34 mm (SD ± 0.82) and 2.1 mm (SD ± 0.75), respectively. There was no significant impact of target depth on targeting accuracy for ND (p = 0.094) or ED (p = 0.187). The mean duration for the planning of one trajectory and for needle positioning were 42 s (SD ± 4) and 64 s (SD ± 7), respectively. In this ex vivo study, the robotic targeting device yielded satisfactory accuracy results at CT slice thicknesses of 1 and 3 mm. This technology may be particularly useful in interventions where the accurate placement of needle-like instruments is required.

The Effect of Implant Length and Bone Quality on the Biomechanical Responses of Dental Implant and Surrounding Bone – A Three-Dimensional Finite Element Analysis

The robustness of dental implant systems in the context of occlusal restoration relies significantly on biomechanical factors associated with the imposition of excessive loads. These factors encompass the macro geometries of the implants, bone qualities, parafunctional oral habits, and specific materials employed. The choice of different implant lengths in different bone qualities may give rise to distinct effects on the way loads are distributed across the interface between the implant and the adjacent bone. As of now, the influence of implant length and bone quality on the surrounding tissues and the stability of implant structure continues to be a matter of debate and uncertainty, particularly in situations involving the possibility of implant failure. This study employed three-dimensional finite element analysis to investigate five distinct implant lengths (4, 6, 10, 13, and 15 mm) in two types of bone quality (type II and III). The bone tissues were characterized through the utilization of computed tomography image datasets and then underwent processing within the SolidWorks software. All geometric configurations were transformed into finite element models, which were subjected to analysis within the ANSYS software. Anisotropic and isotropic properties were attributed to the bone and implant models, respectively. A dynamic occlusal loading quantified at 300 N was applied to the implant body, accompanied by a pre-tension force of 20 N on the screw component. The longer implants exhibited decreased stress magnitudes in type II bone (87.86 – 36.66 MPa) and increased stress magnitudes in type III bone (80.5 – 2128.9 MPa) within the surrounding bone tissue, in comparison to the shorter implants. However, stress within the implant body was generally elevated with the use of longer implants in both bone types (type II: 505.32 – 625.35 MPa; type III: 500.45 – 2186.7 MPa). Irrespective of bone quality, the longer implants predominantly led to lower bone strain levels (type II: 0.006828 – 0.003328; type III: 0.054250 – 0.021678) and overall deformation of the implant-abutment assembly (type II: 0.1458 – 0.1348 μm; type III: 0.1754 – 0.1492 μm) compared to their shorter counterparts. Among all the assessed findings, type III bone displayed a more pronounced adverse impact on the biomechanical responses of the dental implant and neighbouring bone except for the implant stresses under the applied physiological loading.

Modified ResNet152v2: Binary Classification and Hybrid Segmentation of Brain Stroke Using Transfer Learning-Based Approach

Abstract Introduction: The brain is harmed by a medical condition known as a stroke when the blood vessels in the brain burst. Symptoms may appear when the brain’s flow of blood and other nutrients is disrupted. The World Health Organization (WHO) claims that stroke is the leading cause of disability and death worldwide. A stroke can be made less severe by detecting its different warning symptoms early. A brain stroke can be quickly diagnosed using computed tomography (CT) images. Time is passing quickly, although experts are studying every brain CT scan. This situation can cause therapy to be delayed and mistakes to be made. As a result, we focused on using an effective transfer learning approach for stroke detection. Material and methods: To improve the detection accuracy, the stroke-affected region of the brain is segmented using the Red Fox optimization algorithm (RFOA). The processed area is then further processed using the Advanced Dragonfly Algorithm. The segmented image extracts include morphological, wavelet features, and grey-level co-occurrence matrix (GLCM). Modified ResNet152V2 is then used to classify the images of Normal and Stroke. We use the Brain Stroke CT Image Dataset to conduct tests using Python for implementation. Results: Per the performance analysis, the proposed approach outperformed the other deep learning algorithms, achieving the best accuracy of 99.25%, sensitivity of 99.65%, F1-score of 99.06%, precision of 99.63%, and specificity of 99.56%. Conclusions: The proposed deep learning-based classification system returns the best possible solution among all input predictive models considering performance criteria and improves the system’s efficacy; hence, it can assist doctors and radiologists in a better way to diagnose Brain Stroke patients.

Enhancing Airway Assessment with a Secure Hybrid Network-Blockchain System for CT & CBCT Image Evaluation

Our investigation explored the intricacies of airway evaluation through Cone-Beam Computed Tomography (CBCT) and Computed Tomography (CT) images. By employing innovative data augmentation strategies, we expanded our dataset significantly, enabling a more comprehensive analysis of airway characteristics. The utility of these techniques was evident in their ability to yield a diverse array of synthetic images, each representing different airway scenarios with high fidelity. A notable outcome of our study was the effective categorization of the initial image as "Class II" under the Mallampati Classification system. The augmented images further enhanced our understanding by exhibiting a spectrum of airway parameters. Moreover, our approach included training a Recurrent Neural Network (RNN) model on a dataset of CT images. This model, fortified with pseudo-labels created via K-means clustering, showcased its proficiency by accurately predicting airway assessment categories in various test scenarios. These results underscore the model's potential as a tool for swift and precise airway evaluation in clinical settings, marking a significant advancement in medical imaging technologies.

Patient Re-Identification Based on Deep Metric Learning in Trunk Computed Tomography Images Acquired from Devices from Different Vendors

During radiologic interpretation, radiologists read patient identifiers from the metadata of medical images to recognize the patient being examined. However, it is challenging for radiologists to identify “incorrect” metadata and patient identification errors. We propose a method that uses a patient re-identification technique to link correct metadata to an image set of computed tomography images of a trunk with lost or wrongly assigned metadata. This method is based on a feature vector matching technique that uses a deep feature extractor to adapt to the cross-vendor domain contained in the scout computed tomography image dataset. To identify “incorrect” metadata, we calculated the highest similarity score between a follow-up image and a stored baseline image linked to the correct metadata. The re-identification performance tests whether the image with the highest similarity score belongs to the same patient, i.e., whether the metadata attached to the image are correct. The similarity scores between the follow-up and baseline images for the same “correct” patients were generally greater than those for “incorrect” patients. The proposed feature extractor was sufficiently robust to extract individual distinguishable features without additional training, even for unknown scout computed tomography images. Furthermore, the proposed augmentation technique further improved the re-identification performance of the subset for different vendors by incorporating changes in width magnification due to changes in patient table height during each examination. We believe that metadata checking using the proposed method would help detect the metadata with an “incorrect” patient identifier assigned due to unavoidable errors such as human error.

Automated Midline Shift Detection and Quantification in Traumatic Brain Injury: A Comprehensive Review

AbstractTraumatic brain injury (TBI) often results in midline shift (MLS) that is a critical indicator of the severity and prognosis of head injuries. Automated analysis of MLS from head computed tomography (CT) scans using artificial intelligence (AI) techniques has gained much attention in the past decade and has shown promise in improving diagnostic efficiency and accuracy. This review aims to summarize the current state of research on AI-based approaches for MLS analysis in TBI cases, identify the methodologies employed, evaluate the performance of the algorithms, and draw conclusions regarding their potential clinical applicability. A comprehensive literature search was conducted, identifying 15 distinctive publications. The identified articles were analyzed for their focus on MLS detection and quantification using AI techniques, including their choice of AI algorithms, dataset characteristics, and methodology. The reviewed articles covered various aspects related to MLS detection and quantification, employing deep neural networks trained on two-dimensional or three-dimensional CT imaging datasets. The dataset sizes ranged from 11 patients' CT scans to 25,000 CT images. The performance of the AI algorithms exhibited variations in accuracy, sensitivity, and specificity, with sensitivity ranging from 70 to 100%, and specificity ranging from 73 to 97.4%. AI-based approaches utilizing deep neural networks have demonstrated potential in the automated detection and quantification of MLS in TBI cases. However, different researchers have used different techniques; hence, critical comparison is difficult. Further research and standardization of evaluation protocols are needed to establish the reliability and generalizability of these AI algorithms for MLS detection and quantification in clinical practice. The findings highlight the importance of AI techniques in improving MLS diagnosis and guiding clinical decision-making in TBI management.

HRCTCov19-a high-resolution chest CT scan image dataset for COVID-19 diagnosis and differentiation.

Computed tomography (CT) was a widely used diagnostic technique for COVID-19 during the pandemic. High-Resolution Computed Tomography (HRCT), is a type of computed tomography that enhances image resolution through the utilization of advanced methods. Due to privacy concerns, publicly available COVID-19 CT image datasets are incredibly tough to come by, leading to it being challenging to research and create AI-powered COVID-19 diagnostic algorithms based on CT images. To address this issue, we created HRCTCov19, a new COVID-19 high-resolution chest CT scan image collection that includes not only COVID-19 cases of Ground Glass Opacity (GGO), Crazy Paving, and Air Space Consolidation but also CT images of cases with negative COVID-19. The HRCTCov19 dataset, which includes slice-level and patient-level labeling, has the potential to assist in COVID-19 research, in particular for diagnosis and a distinction using AI algorithms, machine learning, and deep learning methods. This dataset, which can be accessed through the web at http://databiox.com , includes 181,106 chest HRCT images from 395 patients labeled as GGO, Crazy Paving, Air Space Consolidation, and Negative.

Robust explanation supervision for false positive reduction in pulmonary nodule detection.

Lung cancer is the deadliest and second most common cancer in the United States due to the lack of symptoms for early diagnosis. Pulmonary nodules are small abnormal regions that can be potentially correlated to the occurrence of lung cancer. Early detection of these nodules is critical because it can significantly improve the patient's survival rates. Thoracic thin-sliced computed tomography (CT) scanning has emerged as a widely used method for diagnosing and prognosis lung abnormalities. The standard clinical workflow of detecting pulmonary nodules relies on radiologists to analyze CT images to assess the risk factors of cancerous nodules. However, this approach can be error-prone due to the various nodule formation causes, such as pollutants and infections. Deep learning (DL) algorithms have recently demonstrated remarkable success in medical image classification and segmentation. As an ever more important assistant to radiologists in nodule detection, it is imperative ensure the DL algorithm and radiologist to better understand the decisions from each other. This study aims to develop a framework integrating explainable AI methods to achieve accurate pulmonary nodule detection. A robust and explainable detection (RXD) framework is proposed, focusing on reducing false positives in pulmonary nodule detection. Its implementation is based on an explanation supervision method, which uses nodule contours of radiologists as supervision signals to force the model to learn nodule morphologies, enabling improved learning ability on small dataset, and enable small dataset learning ability. In addition, two imputation methods are applied to the nodule region annotations to reduce the noise within human annotations and allow the model to have robust attributions that meet human expectations. The 480, 265, and 265 CT image sets from the public Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI) dataset are used for training, validation, and testing. Using only 10, 30, 50, and 100 training samples sequentially, our method constantly improves the classification performance and explanation quality of baseline in terms of Area Under the Curve (AUC) and Intersection over Union (IoU). In particular, our framework with a learnable imputation kernel improves IoU from baseline by 24.0% to 80.0%. A pre-defined Gaussian imputation kernel achieves an even greater improvement, from 38.4% to 118.8% from baseline. Compared to the baseline trained on 100 samples, our method shows less drop in AUC when trained on fewer samples. A comprehensive comparison of interpretability shows that our method aligns better with expert opinions. A pulmonary nodule detection framework was demonstrated using public thoracic CT image datasets. The framework integrates the robust explanation supervision (RES) technique to ensure the performance of nodule classification and morphology. The method can reduce the workload of radiologists and enable them to focus on the diagnosis and prognosis of the potential cancerous pulmonary nodules at the early stage to improve the outcomes for lung cancer patients.

Enhancing facial feature de-identification in multiframe brain images: A generative adversarial network approach.

The collection of head images for public datasets in the field of brain science has grown remarkably in recent years, underscoring the need for robust de-identification methods to adhere with privacy regulations. This paper elucidates a novel deep learning-based approach to deidentifying facial features in brain images using a generative adversarial network to synthesize new facial features and contours. We employed the precision of the three-dimensional U-Net model to detect specific features such as the ears, nose, mouth, and eyes. Results: Our method diverges from prior studies by highlighting partial regions of the head image rather than comprehensive full-head images. We trained and tested our model on a dataset comprising 490 cases from a publicly available head computed tomography image dataset and an additional 70 cases with head MR images. Integrated data proved advantageous, with promising results. The nose, mouth, and eye detection achieved 100% accuracy, while ear detection reached 85.03% in the training dataset. In the testing dataset, ear detection accuracy was 65.98%, and the validation dataset ear detection attained 100%. Analysis of pixel value histograms demonstrated varying degrees of similarity, as measured by the Structural Similarity Index (SSIM), between raw and generated features across different facial features. The proposed methodology, tailored for partial head image processing, is well suited for real-world imaging examination scenarios and holds potential for future clinical applications contributing to the advancement of research in de-identification technologies, thus fortifying privacy safeguards.

Covid-19 detection using modified xception transfer learning approach from computed tomography images

The significance of efficient and accurate diagnosis amidst the unique challenges posed by the COVID-19 pandemic underscores the urgency for innovative approaches. In response to these challenges, we propose a transfer learning-based approach using a recently annotated Computed Tomography (CT) image database. While many approaches propose an intensive data preprocessing and/or complex model architecture, our method focuses on offering an efficient solution with minimal manual engineering. Specifically, we investigate the suitability of a modified Xception model for COVID-19 detection. The method involves adapting a pre-trained Xception model, incorporating both the architecture and pre-trained weights from ImageNet. The output of the model was designed to make the final diagnosis decisions. The training utilized 128 batch sizes and 224x224 input image dimensions, downsized from standard 512x512. No further da processing was performed on the input data. Evaluation is conducted on the 'COV19-CT-DB' CT image dataset, containing labeled COVID-19 and non-COVID-19 cases. Results reveal the method's superiority in accuracy, precision, recall, and macro F1 score on the validation subset, outperforming the VGG-16 transfer model and thus offering enhanced precision with fewer parameters. Furthermore, compared to alternative methods for the COV19-CT-DB dataset, our approach exceeds the baseline approach and other alternatives on the same dataset. Finally, the adaptability of the modified Xception transfer learning-based model to the unique features of the COV19-CT-DB dataset showcases its potential as a robust tool for enhanced COVID-19 diagnosis from CT images.

A hybrid few-shot multiple-instance learning model predicting the aggressiveness of lymphoma in PET/CT images

Background and objectivePatients with aggressive non-Hodgkin lymphoma (NHL) undergo distinct therapy strategies compared with indolent NHL patients. However, it is challenging to estimate NHL aggressiveness based on visual inspection of positron emission tomography (PET) or computed tomography (CT) images. Since diffuse large B-cell lymphoma (DLBCL) and Follicular lymphoma (FL) are the most typical and dominant aggressive and indolent NHL, respectively, this study aims to develop an artificial-intelligence-enabled model to distinguish DLBCL from FL in PET/CT images as the first step to tackle this challenge. MethodsWe propose a hybrid few-shot multiple-instance learning model to predict the aggressiveness of the NHL. First, rotation-based self-supervision learning (SSL) has been employed to train the encoder on a large-scale, publicly available CT image dataset. Second, hybrid instance-level features are obtained for each NHL lesion by combining deep features with the radiomics features from both PET and CT modalities. Third, instance-level features are transformed into bag-level (or patient-level) representations. Finally, bag-level representations are fed into a distance-based classifier through few-shot learning to predict NHL aggressiveness. ResultsOur model achieves an accuracy of 0.751 ± 0.008, a sensitivity of 0.787 ± 0.012, a specificity of 0.715 ± 0.013, an F1-score of 0.753 ± 0.009, and an area under the curve (AUC) of 0.795 ± 0.009 at the bag level. It outperforms the typical counterparts that use the radiomic features, random forest for feature selection, and support vector machines (SVMs) as classifiers. The three counterparts yield accuracies of 0.714 ± 0.023, 0.705 ± 0.008, and 0.698 ± 0.008, respectively. Moreover, settings of the SSL training dataset (Deep lesion) and task (rotation), hybrid CT and radiomic PET features, the pool-layer strategy of maximum, and distance-based classifier generate the best model. ConclusionsA hybrid few-shot multiple-instance learning model can predict lymphoma aggressiveness in PET/CT images and could be a potential tool for determining therapy strategies. Hybrid features and the combination of SSL, few-shot learning, and weakly supervised learning are the two powerful pillars of the model, and these can be expanded to other medical applications with limited samples and incomplete annotations.