Scan Images Research Articles

Low-contrast detectability of photon-counting-detector CT at different scan modes and image types in comparison with energy-integrating-detector CT.

We aim to compare the low-contrast detectability of a clinical whole-body photon-counting-detector (PCD)-CT at different scan modes and image types with an energy-integrating-detector (EID)-CT. We used a channelized Hotelling observer (CHO) previously optimized for quality control purposes. An American College of Radiology CT accreditation phantom was scanned on both PCD-CT and EID-CT with 10 phantom positionings. For PCD-CT, images were generated using two scan modes, standard resolution (SR) and ultra-high-resolution (UHR); two image types, virtual monochromatic images at 70keV and low-energy threshold (T3D); both filtered-back-projection (FBP) and iterative reconstruction (IR) reconstruction methods; and three reconstruction kernels. For each positioning, three repeated scans were acquired for each scan mode, image type, and CTDIvol of 6, 12, and 24mGy. For EID-CT, images acquired from scans (10 positionings × 3 repeats × 3 doses) were reconstructed using the closest counterpart FBP and IR kernels. CHO was applied to calculate the index of detectability () on both scanners. With the smooth Br44 kernel, the of UHR was mostly comparable with that of the SR mode (difference: -11.4% to 8.3%, to 0.956), and the T3D images had a higher (difference: 0.7% to 25.6%) than 70keV images on PCD-CT. Compared with the EID-CT, UHR-T3D of PCD-CT had non-inferior (difference: -2.7% to 12.9%) with IR and non-superior (difference: 0.8% to 11.2%) with FBP using the Br44 kernel. PCD-CT produced higher than EID-CT by 61.8% to 247.1% with the sharper reconstruction kernels. The comparison between PCD-CT and EID-CT was significantly influenced by the reconstruction method and kernel. With a smooth kernel that is typically used in low-contrast detection tasks, the PCD-CT demonstrated low-contrast detectability that was comparable to EID-CT with IR and showed no superiority when using FBP. With the use of sharper kernels, the PCD-CT significantly outperformed EID-CT in low-contrast detectability.

Computer Patient-Specific 3D Modeling and Custom-Made Guides for Revision ACL Surgery.

Revision anterior cruciate ligament reconstruction (ACLR) is a challenging surgery occurring in 3 to 24% of primary reconstructions. A meticulous planning to study the precise size and location of both femoral and tibial bone tunnels is mandatory. The aim of the study was to evaluate the intra- and interoperator differences in the decision-making process between experienced surgeons after they were asked to make preoperative planning for ACL revision reconstruction with the use of both the computed tomography (CT) scan and a three-dimensional (3D)-printed model of the knee. Data collected from 23 consecutive patients undergoing revision of ACLR for graft failure at a single institute between September 2018 and February 2020 were prospectively reviewed. The double-blinded collected data were presented to three board-certificate attending surgeons. Surgeons were asked to decide whether to perform one-stage or two-stage revision ACLR based on the evaluation of the CT scan images and the 3D-printed custom-made models at two different rounds, T0 and T1, respectively, 7 days apart one from the other. Interoperator consensus following technical mistake was 52% at T0 and 56% at T1 using the CT scans, meanwhile concordance was 95% at T0 and 94% at T1 using the 3D models. Concordance between surgeons following new knee injury was 66% at T0 and 70% at T1 using CT scans, while concordance was 96% both at T0 and T1 using 3D models. Intraoperative variability using 3D models was extremely low: concordance at T0 and T1 was 98%. McNemar test showed a statistical significance in the use of 3D model for preoperative planning (p < 0.005). 3D-printed model reliability resulted to be higher compared with CT as intraoperator surgery technique selection was not modified throughout time from T0 to T1 (p < 0.005). The use of 3D-printed models had the most impact when evaluating femoral and tibial tunnels, resulting to be a useful instrument during preoperative planning of revision ACLR between attending surgeons with medium-high workflow.

VER-Net: a hybrid transfer learning model for lung cancer detection using CT scan images

BackgroundLung cancer is the second most common cancer worldwide, with over two million new cases per year. Early identification would allow healthcare practitioners to handle it more effectively. The advancement of computer-aided detection systems significantly impacted clinical analysis and decision-making on human disease. Towards this, machine learning and deep learning techniques are successfully being applied. Due to several advantages, transfer learning has become popular for disease detection based on image data.MethodsIn this work, we build a novel transfer learning model (VER-Net) by stacking three different transfer learning models to detect lung cancer using lung CT scan images. The model is trained to map the CT scan images with four lung cancer classes. Various measures, such as image preprocessing, data augmentation, and hyperparameter tuning, are taken to improve the efficacy of VER-Net. All the models are trained and evaluated using multiclass classifications chest CT images.ResultsThe experimental results confirm that VER-Net outperformed the other eight transfer learning models compared with. VER-Net scored 91%, 92%, 91%, and 91.3% when tested for accuracy, precision, recall, and F1-score, respectively. Compared to the state-of-the-art, VER-Net has better accuracy.ConclusionVER-Net is not only effectively used for lung cancer detection but may also be useful for other diseases for which CT scan images are available.

Integrating Yolo V5 Analysis and KNN to Improve Lung Cancer Detection

Lung cancer is a leading cause of cancer-related mortality globally, emphasizing the urgent need for early detection and accurate diagnosis. This project aims to leverage advanced deep learning techniques, specifically YOLO-v5 (You Only Look Once) for object detection, and the k-Nearest Neighbors (kNN) algorithm for unsupervised learning, to enhance the detection and analysis of lung cancer from CT scan images. YOLO-v5, known for its exceptional speed and accuracy in detecting objects within images, will be used to identify and localize lung nodules, which are potential indicators of lung cancer. Simultaneously, we will employ the kNN algorithm in a novel application of unsupervised learning to cluster CT scan images based on the similarity of detected lung tumors, enabling the identification of patterns and characteristics that may correlate with specific types of lung cancer. This project involves collecting and preprocessing a diverse dataset of CT images annotated with radiologist insights to train the YOLO-v5 model. Subsequently, the kNN algorithm will be applied to perform clustering on the detected tumors. By achieving high accuracy in nodule detection and effectively clustering similar tumors, the system aims to become an invaluable tool for radiologists, providing rapid diagnostic assistance and facilitating a deeper understanding of lung cancer characteristics.

CMC strength and damage prediction considering thermal mechanic coupling effect

Continuous Fiber Reinforced Ceramic Matrix Composites (CMC) confront significant and undeniable elevated thermal gradient. In this study, a periodic RVE geometrical model was simplified based on the XCT scanning images of 2.5D woven CMC, and the model of the thermal structural strength of 2.5D woven CMC under the influence of fine-scale temperature fields was developed. In conjunction with the temperature distribution of the aero-engine turbine working state, ten sets of temperature field conditions were established and the model of thermal structural strength of 2.5D woven CMC under the influence of fine-scale temperature fields was developed. The effects of the overall temperature level, the form of the temperature difference and the size of the temperature difference on the tensile curve, the failure limits, the form of damage evolution, and the distribution of the failure unit at the time of failure of the 2.5D woven CMC were discussed. It was found that the temperature field induced by temperature difference in the thickness direction has a more significant effect on the tensile mechanical properties of the material, which is a key consideration for the structural design process using this material.

Dynamic Survival Prediction Using Sparse Longitudinal Images via Multi-Dimensional Functional Principal Component Analysis

Our work is motivated by predicting the progression of Alzheimer’s disease (AD) based on a series of longitudinally observed brain scan images. Existing works on dynamic prediction for AD focus primarily on extracting predictive information from multivariate longitudinal biomarker values or brain imaging data at the baseline; whereas in practice, the subject’s brain scan image represented by a multi-dimensional data matrix is collected at each follow-up visit. It is of great interest to predict the progression of AD directly from a series of longitudinally observed images. We propose a novel multi-dimensional functional principal component analysis based on alternating regression on tensor-product B-spline, which circumvents the computational difficulty of doing eigendecomposition, and offers the flexibility of accommodating sparsely and irregularly observed image series. We then use the functional principal component scores as features in the Cox proportional hazards model. We further develop a dynamic prediction framework to provide a personalized prediction that can be updated as new images are collected. Our method extracts visibly interpretable images of the functional principal components and offers an accurate prediction of the conversion to AD. We examine the effectiveness of our method via simulation studies and illustrate its application on the Alzheimer’s Disease Neuroimaging Initiative data. Supplementary materials for this article are available online.

Research on an SICM Scanning Image Resolution Enhancement Algorithm.

Scanning ion conductance microscopy (SICM) enables the non-invasive three-dimensional imaging of live cells and other structures in physiological environments. However, when imaging complex samples, SICM faces challenges such as having a low temporal resolution during slow scanning and a reduced signal-to-noise ratio during fast scanning, making it difficult to simultaneously improve both temporal and spatial resolution. To address these issues, this paper proposes an algorithm for enhancing image resolution under high-speed scanning. Firstly, scanning images are preprocessed using a median filtering algorithm to remove the salt-and-pepper noise generated during high-speed scanning. Next, the Canny edge detection algorithm is employed to extract the edges of the image targets. To avoid blurring the edges, the new edge-directed interpolation (NEDI) algorithm is then used to fill the edges, while non-edge areas are filled using bilinear interpolation, thereby enhancing the image resolution. Finally, the peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) are used to analyze the imaging of articular chondrocytes. The results show that under a scanning speed of 480 nm/ms, the proposed algorithm improves the temporal resolution of imaging by 60% compared to traditional 2× resolution imaging, increases the peak signal-to-noise ratio of the scanning images by 7 dB, and achieves a structural similarity of 0.97. Therefore, the proposed algorithm effectively removes noise during high-speed scanning and improves the SICM scanning imaging resolution, thereby avoiding the reduction in temporal resolution when scanning larger resolution samples and effectively enhancing the performance of SICM scanning imaging.

Single-photon lidar depth imaging based on sparsity adaptive matching pursuit method

Photon counting lidar has the ability to detect weak echo photons. However, in imaging applications, a small number of echo signals obtained in a very short acquisition time cannot get an accurate image estimation through the photon arrival time distribution histogram. Therefore, high-precision and fast imaging of an object using a small amount of weak echo photon signals is urgent to be solved. In this work, a sparse adaptive matching pursuit (SAMP) algorithm is proposed to process the echo photons without prior information to obtain the effective photon interval, which improves the reconstruction quality and imaging speed. By constructing a non-coaxial scanning imaging system to detect the face model, it has been experimentally proven that the proposed algorithm can achieve depth and reflectance image estimation under the condition of low power and low acquisition time of 4µs per pixel on average, and the mean absolute error (MAE) value is reduced by about 10 times compared to conventional reconstruction algorithm.

Flexible Sensor for Real-Time Monitoring of Motion Artifacts in Magnetic Resonance Imaging.

In recent years, magnetic resonance imaging has been widely used in the medical field. During the scan, if the human body moves, then there will be motion artifacts on the scan image, which will interfere with the diagnosis and only be found after the end of the scan sequence, resulting in a waste of manpower and resources. However, there is a lack of technology that halts scanning once motion artifacts arise. Here, we designed a real-time monitoring sensor (RMS) to dynamically perceive the movement of the human body and to pause in time when the movement exceeds a certain amplitude. The sensor has an array structure that can accurately sense the position of the human body in real time. The selection of the RMS ensures that there is no additional interference with the scanning results. Based on this design, the RMS can achieve the monitoring function of motion artifact generation.

Modeling and control of two-stage magnetically suspended rotating payload platform

The new generation of remote sensing satellites improves the coverage of remote sensing imaging by improving the orbital surface and using the rotary scanning imaging technology. In this paper, a new technical scheme of rotating payload platform based on disturbance-free payload technology is proposed to adapt to the rotary scanning imaging satellite. The platform adopts the cubic Stewart configuration and uses the Lorentz force magnetic actuator as the actuator to form the satellite support module and the payload into a noncontact two-stage three-body. First, this paper uses Newton-Euler method to establish a six-degree-of-freedom dynamic model for the new rotating payload platform, and establishes a kinematic model by quaternion method. Then, according to the established kinematics and dynamics model, the control scheme of absolute pointing control of payload platform (upper platform) and relative follow-up control of kinetic energy container (middle platform) is designed. Furthermore, the six-degree-of-freedom decoupling control method of the actuator is designed. Based on this, the absolute pointing controller of the payload module and the relative position and attitude calculate controller of the kinetic energy container are designed. Finally, the research of simulation is carried out. The simulation results show that the new rotating payload platform has higher rotation pointing accuracy and stability than the traditional ultra-quiet platform, and the vibration isolation acting on the support module and the vibration absorption acting on the payload are more effective.

Anterior cruciate ligament femoral side retained stump technique reduces enlargement of the femoral bone tunnel after anterior cruciate ligament reconstruction

BackgroundEnlargement of the bone tunnel has become an unavoidable early complication after anterior cruciate ligament (ACL) reconstruction, whether it is a single or double-bundle ACL reconstruction. Preservation of the ACL stump in ACL reconstruction reduces enlargement of the bone tunnel. The purpose of this study was to investigate the question of whether single-bundle ACL reconstruction using the ACL femoral side retained stump technique reduces enlargement of the femoral tunnel.MethodsForty patients who underwent single-bundle reconstruction of the ACL were included in this study. The patients were categorized into a Remnant preservation group (Group R) and the Non-remnant preservation group (Group N). In the Remnant preservation group, a high-flexion femoral side retained stump technique was used intraoperatively for the establishment of the femoral side bone tunnel, and in the Non-remnant preservation group, the conventional femoral positioning method was used (we used a femoral positioning drill for localization and drilling of the femoral bone tunnel), and MRI of the operated knee joints was performed at 6 months postoperatively. We measured the internal diameter of the femoral bone tunnel at 5 mm from the intra-articular outlet of the femoral bone tunnel on an MRI scan image perpendicular to the femoral bone tunnel. The size of the tunnel was compared between the intraoperative drilling of the bone tunnel and the size of the bone tunnel at 6 months postoperatively. Postoperative clinical assessment was Lysholm score.ResultsAfter a 6-month follow-up of 40 patients, the diameter of the femoral tunnel at a distance of 5 mm from the inner opening of the femoral tunnel was 10.96 ± 0.67 mm and 10.11 ± 0.62 mm in patients of group N and group R, respectively, and the difference was statistically significant (P < 0.05).The diameter of the femoral tunnel at 6 months postoperatively in group N and group R compared to the intraoperative bone tunnel increased by 2.58 ± 0.24 mm and 1.94 ± 0.31 mm, and the difference was statistically significant (P < 0.05).The femoral tunnel enlargement rates of group N and group R were 30.94 ± 3.00% and 24.02 ± 5.10%, respectively, and the differences were significant (P < 0.05).ConclusionACL femoral side retained stump technique does not sacrifice the ideal location of the femoral tunnel and is able to preserve the possible benefits of the ACL stump: reduced femoral tunnel enlargement.

Pore morphology changes and quantitative analysis and evaluation of high‐rank coal samples after impact

AbstractThe influencing mechanism of impact loads on the pore morphology of high‐rank coal has been extensively investigated since mining safety is greatly determined by the ensuing state of gas occurrence. In this work, we propose a combined use of a scanning electron microscope and Image Pro Plus software to examine the impact‐induced changes in pore characteristics. Impact loads of 0, 28.46, 51.8, and 58.7 MPa were applied to the coal samples in various directions (parallel/vertical/45° oblique). The results show that the pore connectivity of coal samples is enhanced following the impact load, and the number and diameter of pores are significantly increased on the whole. Surface porosity and roundness of the coal body first rise and then fall as the impact load increases. Following the impact load, surface porosity and roundness of the coal body significantly increase in the vertical and oblique bedding directions, while decreasing in the horizontal bedding direction.

Permeability evolution of fractured coal subject to confining stress and true triaxial stress loading: experiment and mathematical model

The accurate elucidation and prediction of coal permeability evolution under stress loading conditions are crucial for coalbed methane production. In this study, flow experiments were conducted on six cylindrical coal samples and four cubic coal samples under both confining and true triaxial stress loading conditions, respectively. The structure and characteristic parameters of the fractures inside each coal sample were obtained using the computed tomography scanning system and image processing technologies. The coal permeability under both types of loading processes was calculated through the transient pulse method. A mathematical model was developed to assess the evolution of coal permeability under true triaxial loading based on the current true triaxial permeability model and fractal theory. The results revealed that during the confining pressure loading, the coal permeability decreased exponentially with effective stress and was effectively described using the SD model. Additionally, the coal permeability initially rapidly decreased, followed by a gradual decrease, and eventually stabilized at a constant value. Particularly, during the first three loading steps, the fracture aperture and corresponding permeability of the six cylindrical coal samples decreased by ∼51.79%–57.83% and ∼38.06%–42.12%, respectively. However, during the final three loading steps, the fracture aperture and corresponding permeability of the six coal samples decreased by ∼18.26%–23.08% and ∼22.15%–26.93%, respectively. Moreover, owing to the various crossing angles of complex fracture networks with each principal stress, the effect of each principal stress on the coal permeability evolution was highly anisotropic during triaxial stress loading. Particularly, the permeability of the ST1 sample decreased by 43.08%, 14.84%, and 42.08% during the loading of each principal stress. Similarly, the permeability of the ST2 sample decreased by ∼65.74%, 14.29%, and 19.97%. The permeability reductions for the ST3 sample were ∼34.03%, 55.85%, and 10.12%, while those for the ST4 sample were ∼35.97%, 46.51%, and 17.52%. The SD model failed to describe these anisotropic effects. Compared with the SD model, the improved model, based on the current true triaxial permeability model and fractal theory, effectively described the anisotropic effect of each principal stress on the permeability of coal samples with complex fracture networks under triaxial stress conditions.

Biomechanical Effect of Different Posterior Fixation Techniques on Stability and Adjacent Segment Degeneration in Treating Thoracolumbar Burst Fracture With Osteoporosis: A Finite Element Analysis.

Finite element analysis. To investigate the biomechanical effect of four posterior fixation techniques on stability and adjacent segment degeneration in treating thoracolumbar burst fractures with osteoporosis. In terms of stability and adjacent segment degeneration, there remains no consensus or guidelines on the optimal technique for the treatment of thoracolumbar burst fractures in patients with osteoporosis. Images of CT scans were imported into MIMICS and further processed by Geomagic to build three-dimensional models of the T10-L5 region. A v-shaped osteotomy was performed on the L1 vertebral body to simulate a burst fracture in the setting of osteoporosis. Subsequently, four fixation techniques were designed using SolidWorks software. Range of motion (ROM) of the global spine, ROM distribution, ROM of adjacent segment, Von Mises stress on adjacent intervertebral disks, and facet joints were analyzed. Among the four groups, the cortical bone screw fixation (CBT) showed the highest global ROM at 1.86°, while long-segmented pedicle screw fixation (LSPS) had the lowest global ROM at 1.25°. The LSPS had the smallest percentage of ROM of fractured vertebral body to fixed segment at 75.04%, suggesting the highest stability after fixation. The maximum ROM of the adjacent segment was observed in the CBT at 1.32°, while the LSPS exhibited the smallest at 0.89°. However, the LSPS group experienced larger maximum stress on the adjacent intervertebral disks (9.60MPa) and facet joints (3.36MPa), indicating an increasing risk of adjacent segment disease. LSPS provided the greatest stability, while CBT provided the smallest amount of stability. However, the elevated stress on adjacent intervertebral disks and facet joints after LSPS fixation increased the possibility of adjacent segment degeneration. Cement-augmented pedicle screw fixation (CAPS) and combined cortical bone screw and pedicle screw fixation (CBT-PS) demonstrated significant biomechanical advantages in providing moderate fixation strength while reducing stress on the intervertebral disks and facet joints.

Thermal design and experimental verification of the space ultraviolet wide field of view multi-channel scanning imager

In order to ensure the optimal imaging quality of the space ultraviolet wide field of view multi-channel scanning imager, it is imperative to conduct a comprehensive investigation into the thermal control system. First, we shall present the structure and performance index of the imager. The determination of the heat dissipation surface's position follows an analysis of the imager's exterior heat discharge. Subsequently, the thermal control design for every individual component of the imager is determined, considering the challenges associated with thermal management. The accuracy of the initial thermal design is verified by doing thermal simulation studies. Subsequently, the thermal equilibrium test was performed. The test results reveal that the temperature of the lens assembly varied between 20.0 °C and 25.6 °C, the horizontal bracket ranged from 19.7 °C to 21.7 °C, the vertical bracket exhibited temperatures between 11.2 °C and 15.4 °C, the U-frame experienced temperatures ranging from 15.1 °C to 17.4 °C, the motor showed temperatures between 6.6 °C and 26.3 °C, the detector recorded temperatures between 21.4 °C and 25.9 °C, and the high voltage module exhibited temperatures ranging from 8.9 °C to 19.9 °C. The temperature of each component is within the designated thermal control range. Extensive evidence has been presented to demonstrate the effectiveness and precision of thermal design.

Computerized tomography on the 5th postoperative day helps distinguish grade C from grade B pancreatic fistula after pancreaticoduodenectomy for periampullary cancer

ABSTRACT Background Grade C pancreatic fistula (PF) is the most dangerous complication after pancreaticoduodenectomy (PD) and had not been identified on the early post-operative day yet. In this study, we aimed to introduce a new approach to distinguish grade C from grade B PF by using features on the 5th post-operative day’s computerized tomography (CT). Materials and Methods A bi-centric prospective cohort study was conducted from August 2021 to October 2023 at the University Medical Center (UMC) and Cho Ray Hospital, Viet Nam. Patients who diagnosed pancreatic fistula, underwent a CT scan on the 5th postoperative day were followed - up within 90 days after PD. The characteristics of CT scan images were analyzed to differentiate both grades of PF. Results During 26 months, 77 patients with periampullary cancer underwent PD and met the criteria of the study (46 males, mean age: 57.1 ± 13.7). Of these, 52 patients had biochemical PF and 25 had CRPF (18 grade B PF and 7 grade C PF). Specifications on the 5th postoperative day’s CT scan that can discriminate grade C from grade B included: fluid diffusing throughout abdomen, retroperitoneal abscess, hematoma beneath the liver, size of the fluid collection next to the pancreas (p = 0.031) and size of pancreatojejunostomy anastomosis dehiscence (p = 0.045). Conclusions Features on CT scan could be used to identify grade C from the 5th postoperative day. This result can help surgeons to make a plan for carefully follow-up those patients and indicate re-operation at an early stage.

EfficientNet Transfer Learning Approach for Multi-Class Brain Tumor Classification

- It can be difficult for radiologists to accurately diagnose brain tumors in the medical field. When radiologists manually identify tumors, it can result in incorrect treatment planning and medical errors. The Miami Neuroscience Center states that there are 120 different kinds of brain tumors that can impact an individual's brain. Brain tumors that are most common and dangerous include glioma, meningioma, and pituitary tumors. In this work, we developed an online diagnostic tool to accurately classify brain tumors, including Astrocytoma, Glioma, Meningioma, Neurocytoma and Pituitary. Additionally, our program is capable of classifying a normal brain. We added CT and MRI scan images to improve cross-modality and model capability. We employed the CNN based neural network architecture known as EfficientNet. With this model, we achieved over 93.4% accuracy. Our application will help the radiologist identify tumors quickly and promote better treatment planning. Key words: Deep Learning, Transfer Learning, CNN, MRI, CT, Multi-Class Classification.

Colon and lung cancer classification from multi-modal images using resilient and efficient neural network architectures

Automatic classification of colon and lung cancer images is crucial for early detection and accurate diagnostics. However, there is room for improvement to enhance accuracy, ensuring better diagnostic precision. This study introduces two novel dense architectures (D1 and D2) and emphasizes their effectiveness in classifying colon and lung cancer from diverse images. It also highlights their resilience, efficiency, and superior performance across multiple datasets. These architectures were tested on various types of datasets, including NCT-CRC-HE-100K (set of 100,000 non-overlapping image patches from hematoxylin and eosin (H&E) stained histological images of human colorectal cancer (CRC) and normal tissue), CRC-VAL-HE-7K (set of 7180 image patches from N = 50 patients with colorectal adenocarcinoma, no overlap with patients in NCT-CRC-HE-100K), LC25000 (Lung and Colon Cancer Histopathological Image), and IQ-OTHNCCD (Iraq-Oncology Teaching Hospital/National Center for Cancer Diseases), showcasing their effectiveness in classifying colon and lung cancers from histopathological and Computed Tomography (CT) scan images. This underscores the multi-modal image classification capability of the proposed models. Moreover, the study addresses imbalanced datasets, particularly in CRC-VAL-HE-7K and IQ-OTHNCCD, with a specific focus on model resilience and robustness. To assess overall performance, the study conducted experiments in different scenarios. The D1 model achieved an impressive 99.80 % accuracy on the NCT-CRC-HE-100K dataset, with a Jaccard Index (J) of 0.8371, a Matthew's Correlation Coefficient (MCC) of 0.9073, a Cohen's Kappa (Kp) of 0.9057, and a Critical Success Index (CSI) of 0.8213. When subjected to 10-fold cross-validation on LC25000, the D1 model averaged (avg) 99.96 % accuracy (avg J, MCC, Kp, and CSI of 0.9993, 0.9987, 0.9853, and 0.9990), surpassing recent reported performances. Furthermore, the ensemble of D1 and D2 reached 93 % accuracy (J, MCC, Kp, and CSI of 0.7556, 0.8839, 0.8796, and 0.7140) on the IQ-OTHNCCD dataset, exceeding recent benchmarks and aligning with other reported results. Efficiency evaluations were conducted in various scenarios. For instance, training on only 10 % of LC25000 resulted in high accuracy rates of 99.19 % (J, MCC, Kp, and CSI of 0.9840, 0.9898, 0.9898, and 0.9837) (D1) and 99.30 % (J, MCC, Kp, and CSI of 0.9863, 0.9913, 0.9913, and 0.9861) (D2). In NCT-CRC-HE-100K, D2 achieved an impressive 99.53 % accuracy (J, MCC, Kp, and CSI of 0.9906, 0.9946, 0.9946, and 0.9906) with training on only 30 % of the dataset and testing on the remaining 70 %. When tested on CRC-VAL-HE-7K, D1 and D2 achieved 95 % accuracy (J, MCC, Kp, and CSI of 0.8845, 0.9455, 0.9452, and 0.8745) and 96 % accuracy (J, MCC, Kp, and CSI of 0.8926, 0.9504, 0.9503, and 0.8798), respectively, outperforming previously reported results and aligning closely with others. Lastly, training D2 on just 10 % of NCT-CRC-HE-100K and testing on CRC-VAL-HE-7K resulted in significant outperformance of InceptionV3, Xception, and DenseNet201 benchmarks, achieving an accuracy rate of 82.98 % (J, MCC, Kp, and CSI of 0.7227, 0.8095, 0.8081, and 0.6671). Finally, using explainable AI algorithms such as Grad-CAM, Grad-CAM++, Score-CAM, and Faster Score-CAM, along with their emphasized versions, we visualized the features from the last layer of DenseNet201 for histopathological as well as CT-scan image samples. The proposed dense models, with their multi-modality, robustness, and efficiency in cancer image classification, hold the promise of significant advancements in medical diagnostics. They have the potential to revolutionize early cancer detection and improve healthcare accessibility worldwide.

Morphometric Study of Lumbar Vertebral Pedicles by Using Computed Tomography among Adults in Kano, Nigeria

Abstract Context: Precise anatomical knowledge of lumbar pedicle dimensions is necessary to make a clinical diagnosis of lower back pain, and any structural deviation of the pedicles may result in interference with the weight transmission mechanism and compression of neural structures. Aim: To evaluate the morphometric dimensions of lumbar vertebral pedicles among normal adults in Kano by using computerized tomography scan images. Materials and Method: This was a hospital-based cross-sectional study that was conducted in the radiology department of Aminu Kano Teaching Hospital (AKTH). This study was conducted over 1 year and included 110 adults aged 18–75 years involving both sexes in Kano, using their CT scan images from L1 to L5, following IRB ethics approval. Each vertebral pedicle was measured in transverse and sagittal planes. The measurements included the pedicle width (PW), sagittal pedicle height (SPH), and the pedicle chord length (PCL) of individual pedicles. The mean and standard deviations for each measurement parameter were determined, and parametric correlation tests were performed to determine the relationships between age, sex, and ethnicity. Statistical Analysis Used: Data analysis was performed using the statistical software package IBM SPSS Statistics version-23 for Windows. The significance level was considered at P &lt; 0.05. Results: Mean PW size increased craniocaudally along the lumbar spine from L1 (7.49 ± 1.52 mm) to L5 (14.31 ± 1.93 mm). A gradual decrease in mean SPH sizes from L1 (14.53 ± 1.32 mm) to L5 (11.73 ± 1.38 mm) was noted, while mean PCL showed an irregular “U” pattern, with the smallest size at L4 (44.63 ± 3.42 mm) and highest at L2 (46.80 ± 3.85 mm). All measured dimensions were significantly higher in the right pedicles and among male subjects (P &lt; 0.05). Age and ethnicity did not consistently affect pedicle dimensions (P &gt; 0.05). Conclusion: This study observed right pedicle dimensions and male subjects as having significantly larger pedicle diameters, with age and ethnicity not having a significant effect on pedicle sizes across all vertebral levels.

Three-dimensional Hemifacial Microsomia Classification with New Subtypes Based on the Pruzansky and Kaban Classification.

Various classifications of hemifacial microsomia (HFM) have been described previously. Although some of these classifications are used widely, others use external outlines of reference organs, even in three-dimensional (3D) images. The purpose of this study was to investigate the 3D characteristics of the mandibular condyle in HFM and to update the Pruzansky and Kaban classification as a 3D classification. Fifty-three patients with HFM were classified according to the Pruzansky and Kaban classification (type I, IIA, IIB, and III) using computed tomographic scan images. 3D images of the mandible were isolated, and the 3D characteristics were observed; furthermore, the angle of inclination of the mandibular condyle was measured in 3D. Subtypes of retroflexed mandibular condyle in 3D were observed in the Pruzansky and Kaban classification type IIA and IIB, termed as type IIAβ (33.3% in type IIA) and type IIBβ (100% in type IIB). Although some differences were observed in the inclination of the mandibular condyle between the control and the affected sides in type I and IIAα, multiple differences were observed in type IIAβ and IIBβ. To the best of our knowledge, this is the first report that identified the retroflexed mandibular condyle as subtypes type IIAβ and IIBβ. Notably, this could not be identified in the two-dimensional images (lateral cephalogram) yet. We proposed to update the Pruzansky and Kaban classification as a 3D classification with a new 3D subtype. The angle of the retroflexed mandibular condyle may predict mandibular growth in HFM.