CT Scan Images Research Articles

Machine Learning Based Framework for Lung Cancer Detection and Image Feature Extraction Using VGG16 with PCA on CT-Scan Images

Machine Learning Based Framework for Lung Cancer Detection and Image Feature Extraction Using VGG16 with PCA on CT-Scan Images

Deep learning-based Covid-19 diagnosis: a thorough assessment with a focus on generalization capabilities

The Covid-19 pandemic has significantly spurred the development of deep learning (DL) models for the pathology automatic diagnosis based on CT scan images. However, the assumption about the generalization of the proposed models remains to be assessed and shown for concrete clinical use. In this work, we have investigated the real value of widely used public datasets for the elaboration of DL models that are dedicated to automatic diagnosis of Covid-19 using CT scans. We have collected various international public datasets from 13 countries. Different Convolutional Neural Networks (CNNs) have been trained and their performances carefully assessed. Two evaluations have been conducted: (1) an internal evaluation following a cross-validation procedure, and (2) an external evaluation on real patients coming from new and different sources. The objective is to assess the generalization capabilities considering real-world conditions: different acquisition conditions, devices and configurations. Three families from the most effective CNN models have been selected (ResNet, DenseNet and EfficientNet). These have been fine-tuned, evaluated and used within a training methodology based on transfer learning. The most effective models have been further customized in order to create new models that are dedicated to the task at hand. These models have significantly improved the diagnosis performance.

Robot-assisted technique can achieve accurate screw placement in four-corner fusion and reduce operative difficulty: a cadaver study

BackgroundThe purpose of this study is to explore the feasibility and accuracy of a robot-assisted technique in four-corner fusion compared with traditional freehand operation.MethodsTwenty cadaver specimens were randomly assigned to the robot-assisted group and freehand groups. Three screws were placed percutaneously to fix the capitate-lunate joint, lunate-triquetrum joint, and triquetrum-hamate-capitate joint in each specimen by robot-assisted or freehand technique. The offset between the actual and planned screw positions was determined by merging the images of intraoperative and postoperative CT scans in the robot-assisted group. The centrality of the screw, time-consuming, drilling attempts, and radiation exposure were compared between the two groups.ResultsThe mean offset between the actual and planned screw position was 1.09 (SD: 0.56) mm. The offset at the start point of the screw was significantly lower than that at the endpoint. There was no significant difference in the centrality of the screws, surgical time between the two groups. The number of drilling attempts and the radiation dose received by surgeons were significantly lower in the robot-assisted group.ConclusionsAlthough there was no significant difference in screw centrality between the two groups, the slight offset between the actual and planned screw positions confirmed the feasibility of the robot-assisted technique in four-corner fusion. The robot-assisted technique has advantages in reducing the difficulty of surgery and protecting the surgeon from exposure to large doses of radiation.

Non-small cell lung cancer detection through knowledge distillation approach with teaching assistant.

Non-small cell lung cancer (NSCLC) exhibits a comparatively slower rate of metastasis in contrast to small cell lung cancer, contributing to approximately 85% of the global patient population. In this work, leveraging CT scan images, we deploy a knowledge distillation technique within teaching assistant (TA) and student frameworks for NSCLC classification. We employed various deep learning models, CNN, VGG19, ResNet152v2, Swin, CCT, and ViT, and assigned roles as teacher, teaching assistant and student. Evaluation underscores exceptional model performance in performance metrics achieved via cost-sensitive learning and precise hyperparameter (alpha and temperature) fine-tuning, highlighting the model's efficiency in lung cancer tumor prediction and classification. The applied TA (ResNet152) and student (CNN) models achieved 90.99% and 94.53% test accuracies, respectively, with optimal hyperparameters (alpha = 0.7 and temperature = 7). The implementation of the TA framework improves the overall performance of the student model. After obtaining Shapley values, explainable AI is applied with a partition explainer to check each class's contribution, further enhancing the transparency of the implemented deep learning techniques. Finally, a web application designed to make it user-friendly and classify lung types in recently captured images. The execution of the three-stage knowledge distillation technique proved efficient with significantly reduced trainable parameters and training time applicable for memory-constrained edge devices.

Evaluating process parameters of SLS part of polyamide-12 recycled powder using RSM

Abstract Selective laser sintering (SLS) is a highly flexible rapid prototyping (RP) technology which can use a wide range of powdered materials to generate high quality complex geometries directly from digital models. The SLS process uses a laser to sinter selected areas of powdered materials to produce solid objects layer by layer, according to the 2D data obtained from CT scan or MRI imaging techniques. This paper presents research work for evaluating the effect of process parameters on impact toughness and surface roughness of the part manufactured on Farsoon SS402P SLS from recycled ageing FS 3300PA which is PA-12 based powder and standard operating parameters using Response Surface Methodology (RSM) design of experiments. Impact tests have been performed as per the ASTM D6110 standard using recommendations from ISO 179-2. According to the results, effect of process parameters are successfully analyzed by employing impact toughness and surface roughness tests. Based on the results of the ANOVA analysis, impact toughness increases with increasing laser power (4.467 j/cm2 at 72.5 watts). Similarly, it is also clear from the results that the region with the lowest laser power, high layer thickness, and 90-degree orientation had the lowest surface roughness (3.46 μm).

Using Decision Tree With First and Second-Order Statistical Feature Extraction for Classification of Lung Cancer

<p>The classification of CT-Scan images on images with lung cancer and normal lung has been done by improving the image quality of the median and Gabor filters, extraction of first and second-order statistical features, and decision tree classification. The data used comes from LIDC-IDRI as much as 100 training data and 40 test data. The median filter removes noise without removing edges in the image. A Gabor filter is used to facilitate texture analysis on the image. At the feature extraction stage, statistical variations of the first order, second order statistics and the merging of first and second-order statistics. The best results obtained at the testing stage are program designs with variations of feature extraction combining first and second-order statistics. The level of accuracy obtained is 97.5%, with a sensitivity of 100% and a specificity of 95%.</p>

CTRing: An R package to extract wood density profiles from computed tomography images of discs and logs

Accurately determining the position of pith and accessing tree-ring density profiles, including intra-ring variations, is important for both the forest industry and dendroclimatology. Although several available methods exist for acquiring this information, such as X-ray computed tomography (CT), micro-CT, and X-ray films, the availability of open-source programs for extracting data remains limited. The CTRing package in the R environment integrates a series of functions to detect precisely the pith and tree-ring boundaries and generate tree-ring density profiles using CT images of tree cross sections. Before processing, grey values are transformed into density using a calibration function. Pith position is then detected by combining an adapted Hough Transform method and a one-dimensional edge detector. Tree-ring profiles along the pith-to-bark path of interest are inspected visually, and tree-ring boundaries can be easily added or removed manually via a graphical user interface. After correcting for tree-ring boundaries, the inflection points of a 3rd-degree polynomial obtained from density profiles are used to delimit the earlywood–latewood transition. We tested this package using 60 CT-scanned images of white spruce (Picea glauca (Moench) Voss) discs collected at various tree heights (0%, 25%, 50% and 75% of the total tree height as well as at 1.3m). The pith detection function had an average mean error of 0.72mm with 95% of the automatically detected pith locations that differed by less than 2mm from their manually located positions. Error decreased toward the apex of the tree. The functions of the CTRing package are flexible and can be easily implemented or adapted. The package could also be used with simple images of discs to obtain ring-width time series; however, this use must be evaluated further. Future work with this package involves assessing the use of low-quality images and ring-porous species.

Coronary plaque volume burden and composition in patients with non-hemodynamically significant coronary artery plaques and systemic inflammation

Abstract Introduction Atherosclerotic coronary disease is a leading cause of morbidity and mortality worldwide. Systemic inflammation can affect plaque composition and thus increase cardiovascular risk. Identifying plaque characteristics in these patients could be useful to learn how inflammation (identified by easy to access parameters, like high ferritin and CRP) affects plaque composition in these challenging patients. Purpose We aimed to evaluate the association of systemic inflammation identified by high ferritin and/or high C-reactive protein (CRP) with coronary plaque volume and composition in patients with non-hemodynamically significant coronary artery plaques (stenosis <50%). Methods We prospectively recruited 109 patients referred for noninvasive CT angiography (from November 2023 to January 2024) with suspected coronary artery disease. 20 had totally/partially non-calcified, non-significant coronary plaques (stenosis <50%). Plaque volume was measured. Plaque components were stablished as calcified (>200 UH), lipidic (<60 UH) and fibrous (60-200 UH). Systemic inflammation was identified when ferritin was >300 ng/mL and/or C-reactive protein CRP was >5 mg/L, according to our laboratory high normal limits. CT scan images were acquired with a 128 detectors dual-layer CT scan with prospective-retrospective/synchronization. Intravascular iodine-based contrast agent was administered. Results Mean age was 60 ± 11 years old and 75% of patients were men. 55% had high blood pressure, 65% were current or previous smokers and 40% had dyslipidemia (35% received statins). 10% were diabetic. There were no difference between groups for these characteristics. We analyzed 33 plaques. Fibrous plaque volume was found to be significantly higher in patients with high ferritin (66.9 vs 32.5 mm3; p = 0.04) and in patients with high CRP (78.7 vs 36.4 mm3; p = 0.05). The percentage of fibrous plaque tended to be higher in patients with higher ferritin and CRP (54.9 vs 38.9; p = 0.06 and 60.6 vs 40.7%; p = 0.07, respectively). Conclusions Systemic inflammation affecting coronary plaque composition in non-hemodynamically significant coronary atherosclerotic disease can by identified by high ferritin and high CRP levels. It is associated with higher fibrous plaque volumes. This finding suggest that systemic inflammation can associate more vulnerable plaques and might benefit from more aggressive management of non-hemodynamically significant atherosclerosis.TableFigure

The Effect of the COVID-19 Pandemic on the Working Conditions and Efficiency of Radiographers at Ahvaz Jundishapur University of Medical Sciences' Radiology and CT Scan Imaging Centers

Background: Radiographers are highly skilled imaging technology users who work directly with COVID-19 cases. Objectives: This study aims to assess the impact of COVID-19 on the working conditions and efficiency of radiographers and their level of satisfaction with personal protective equipment (PPE), cleaning supplies, and facilities during the COVID-19 epidemic. Methods: A prospective observational study was conducted among radiographers registered in the radiology and CT scan imaging centers of Ahvaz Jundishapur University of Medical Sciences and some private centers between June and October 2022. A questionnaire with 28 multiple-choice questions was used to collect information. The questionnaire consisted of two sections; the first part contained demographic information. In the second part, participants answered questions about changes in working conditions, such as changes in stress level, work efficiency, the need for more rest, imaging time, use of equipment, and personal protective equipment at work, based on their experiences and observations during the COVID-19 pandemic. The data analysis was performed using Statistical Product Service Solutions (SPSS) software, version 26.0, with statistical significance assumed at a P-value < 0.05. Results: One hundred and fifty radiographers completed the questionnaire. The mean age and mean work experience of participants were 33.2 ± 8.3 years and 9 ± 7.2 years, respectively. The results showed that most participants of both sexes agreed that they needed more rest (P-value < 0.05); their work efficiency had not decreased significantly during the COVID-19 epidemic. The stress level of participants had significantly increased compared to before the epidemic. There were significant differences between government-owned hospitals and private radiology centers regarding the presence of a COVID-19-related protocol, the compulsory use of masks, the existence of decontamination instructions, the type of air decontamination, the contact with imaging equipment manufacturers, the timing of decontamination, and the type of decontamination applied to the imaging equipment (P-value < 0.05). Conclusions: This study demonstrated that the reduction in work efficiency among radiographers was not significant compared to before the COVID-19 pandemic. According to the study's results, radiographers' working conditions underwent numerous adjustments over the course of the COVID-19 epidemic.

A clinical comparison of a digital versus conventional design methodology for transtibial prosthetic interfaces

A transtibial prosthetic interface typically comprises a compliant liner and an outer rigid socket. The preponderance of today’s conventional liners are mass produced in standard sizes, and conventional socket design is labor-intensive and artisanal, lacking clear scientific rationale. This work tests the clinical efficacy of a novel, physics-based digital design framework to create custom prosthetic liner-socket interfaces. In this investigation, we hypothesize that the novel digital approach will improve comfort outcomes compared to a conventional method of liner-socket design. The digital design framework generates custom transtibial prosthetic interfaces starting from MRI or CT image scans of the residual limb. The interface design employs FEA to simulate limb deformation under load. Interfaces are fabricated for 9 limbs from 8 amputees (1 bilateral). Testing compares novel and conventional interfaces across four assessments: 5-min walking trial, thermal imaging, 90-s standing pressure trial, and an evaluation questionnaire. Outcome measures include antalgic gait criterion, skin surface pressures, skin temperature changes, and direct questionnaire feedback. Antalgic gait is compared via a repeated measures linear mixed model while the other assessments are compared via a non-parametric Wilcoxon sign-rank test. A statistically significant (\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$P=0.032$$\\end{document}) decrease in pain is demonstrated when walking on the novel interfaces compared to the conventional. Standing pressure data show a significant decrease in pressure on novel interfaces at the anterior distal tibia (\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$P = 0.012$$\\end{document}), with no significant difference at other measured locations. Thermal results show no statistically significant difference related to skin temperature. Questionnaire feedback shows improved comfort on novel interfaces on posterior and medial sides while standing and the medial side while walking. Study results support the hypothesis that the novel digital approach improves comfort outcomes compared to the evaluated conventional method. The digital interface design methodology also has the potential to provide benefits in design time, repeatability, and cost.

The Effect of Tube Voltage Variations on CT Scan Image Quality at Prof. I G. N. G. Ngoerah General Hospital, Denpasar, Indonesia

Aims: Knowing the SNR and CNR values obtained from the image results that have been varied in the tube voltage. Knowing the effect of variations in the tube voltage value on the quality of the CT scan image that will be produced. Place and Duration of Study: Radiology installation of Prof. I G. N. G. Ngoerah General Hospital, Denpasar from May 2024 to August 2024. Methodology: This study used a Canon Aquilion LB CT Scanner to scan a TOS phantom containing 6 materials (polypropylene, nylon, acrylic, derlin, air, and water) with varying tube voltages (80 kV, 100 kV, 120 kV, 135 kV). The data were analyzed using PSPP and Excel to calculate SNR and CNR, and statistical tests for normality, Pearson correlation, and simple regression were performed. Results: The image data analyzed includes the average value of the object ROI, the average ROI background, and the standard deviation of the background. The data is grouped based on the variation of the tube voltage, the average value and standard deviation of the ROI on each material, and the standard deviation of the background. The results obtained for the relationship between the variation of the tube voltage and the SNR value are that the tube voltage affects the SNR value by 94,02% in polypropylene, 97,79% in nylon, 97,71% in acrylic, 96,42% in derlin, and 94,22% in air. And for the relationship between the tube voltage and the CNR value, the tube voltage affects the CNR value by 93,65% in polypropylene, 97,35% in nylon, 96,81% in acrylic, 95,62% in derlin, and 94,65% in air. Conclusion: Selecting the right tube voltage is important to optimize CT scan image quality, with 120 kV and 135 kV recommended for organs such as the kidney, liver, and lungs as they produce better SNR and CNR. Organs with significant attenuation differences show increased SNR and CNR, while soft tissues such as fat experience decreased contrast at high voltages. The use of tube voltage must be adjusted to diagnostic needs and evaluated through quality control according to BAPETEN regulation Number 2 of 2018.

Analysis of the Relationship between Symptoms, Clinical Diagnosis, and Findings of CT-Scan Images to Pictures of Vertebrobasilar Artery Insufficiency

Vertebrobasilar artery insufficiency (VBI) occurs when the blood flow to the posterior circulation of the brain, supplied by the vertebral arteries, becomes inadequate. Frequently, VBI is detected accidentally through CT-Scan in patients who do not show symptoms directly related to this condition. This study highlights the importance of comprehensive evaluation and early screening to improve VBI management. The main objective of this research is to analyze the relationship between clinical symptoms, clinical diagnosis, and CT-Scan findings to enhance diagnostic accuracy and develop more effective treatment strategies. A cross-sectional design was used in this study, with samples obtained from the medical records of patients diagnosed with Vertebrobasilar Artery Insufficiency at Bethesda Hospital Yogyakarta between January and April 2023. The data were analyzed using the chi-square test and processed with SPSS software. The patient population was dominated by elderly individuals (73%) and males (60.5%), although these differences were not statistically significant.The findings suggest that VBI symptoms vary based on age, with older patients more likely to exhibit stroke and hemiparesis, while younger patients typically experience headaches and vertigo. Screening for risk factors like cholesterol and blood sugar levels is crucial for patients over 45, whereas younger patients may require additional tests to assess cardiometabolic or hypercoagulation risks. Early detection through age-appropriate screening is vital for preventing serious complications and improving patient outcomes. These findings imply that systematic screening can help identify undiagnosed VBI cases, allowing for timely and more effective medical intervention.

Prediction of Transport and Deposition of Porous Particles in the Respiratory System Using Eulerian-Lagrangian Approach.

Deep lung delivery is crucial for respiratory disease treatment. Although nano and submicron particles exhibited a good deposition efficiency in deep regions of the lung, powder nonuniformity and particle agglomeration reduce their efficiency. Inhalation of porous particles (PPs) can overcome the mentioned challenges due to their larger size and low-density. The present study numerically investigates the deposition and penetration efficiency of orally inhaled PPs. A revised drag coefficient was implemented for PP transport. A realistic mouth-throat to the fifth generation of the lung was reconstructed from CT-scan images. A dilute suspension of uniformly distributed particles was considered at three inhalation flow rates (15, 30, and 45 L/min). Governing equations of the flow field and particle transport are solved using an Eulerian-Lagrangian approach. The results demonstrate that inhaling PPs significantly reduces the total and regional deposition of particles. There was also a critical porosity value under moderate and high inhalation flow rates for large PPs. Below this critical value, PP deposition efficiency substantially decreases. Additionally, it was also found that under low inhalation flow rates, the impact of porosity value is negligible. Almost 95% of the PPs penetrate the lower branches. These findings provide particle engineers and pharmaceutics with profound insight into developing novel inhalation techniques and drug delivery methods for deep lung delivery.

Integrated ensemble CNN and explainable AI for COVID-19 diagnosis from CT scan and X-ray images

In light of the ongoing battle against COVID-19, while the pandemic may eventually subside, sporadic cases may still emerge, underscoring the need for accurate detection from radiological images. However, the limited explainability of current deep learning models restricts clinician acceptance. To address this issue, our research integrates multiple CNN models with explainable AI techniques, ensuring model interpretability before ensemble construction. Our approach enhances both accuracy and interpretability by evaluating advanced CNN models on the largest publicly available X-ray dataset, COVIDx CXR-3, which includes 29,986 images, and the CT scan dataset for SARS-CoV-2 from Kaggle, which includes a total of 2,482 images. We also employed additional public datasets for cross-dataset evaluation, ensuring a thorough assessment of model performance across various imaging conditions. By leveraging methods including LIME, SHAP, Grad-CAM, and Grad-CAM++, we provide transparent insights into model decisions. Our ensemble model, which includes DenseNet169, ResNet50, and VGG16, demonstrates strong performance. For the X-ray image dataset, sensitivity, specificity, accuracy, F1-score, and AUC are recorded at 99.00%, 99.00%, 99.00%, 0.99, and 0.99, respectively. For the CT image dataset, these metrics are 96.18%, 96.18%, 96.18%, 0.9618, and 0.96, respectively. Our methodology bridges the gap between precision and interpretability in clinical settings by combining model diversity with explainability, promising enhanced disease diagnosis and greater clinician acceptance.

Measurements of Stray Radiation from Medical Devices in Hospitals

The main objective of this study was to look for the electromagnetic interference produced by medical devices in hospitals and to evaluate its impacts on the proper functions of such devices. However, the present study encountered other important problems with regard to CT scanner and X-ray Imaging systems. Such problems demand fast responses in order to secure health safety for the patients and the operators of such systems. ORIGINAL RESEARCH ARTICLE | Oct. 23, 2024

Lightweight Advanced Deep Neural Network (DNN) Model for Early-Stage Lung Cancer Detection.

Background: Lung cancer, also known as lung carcinoma, has a high mortality rate; however, an early prediction helps to reduce the risk. In the current literature, various approaches have been developed for the prediction of lung carcinoma (at an early stage), but these still have various issues, such as low accuracy, high noise, low contrast, poor recognition rates, and a high false-positive rate, etc. Thus, in this research effort, we have proposed an advanced algorithm and combined two different types of deep neural networks to make it easier to spot lung melanoma in the early phases. Methods: We have used WDSI (weakly supervised dense instance-level lung segmentation) for laborious pixel-level annotations. In addition, we suggested an SS-CL (deep continuous learning-based deep neural network) that can be applied to the labeled and unlabeled data to improve efficiency. This work intends to evaluate potential lightweight, low-memory deep neural net (DNN) designs for image processing. Results: Our experimental results show that, by combining WDSI and LSO segmentation, we can achieve super-sensitive, specific, and accurate early detection of lung cancer. For experiments, we used the lung nodule (LUNA16) dataset, which consists of the patients' 3D CT scan images. We confirmed that our proposed model is lightweight because it uses less memory. We have compared them with state-of-the-art models named PSNR and SSIM. The efficiency is 32.8% and 0.97, respectively. The proposed lightweight deep neural network (DNN) model archives a high accuracy of 98.2% and also removes noise more effectively. Conclusions: Our proposed approach has a lot of potential to help medical image analysis to help improve the accuracy of test results, and it may also prove helpful in saving patients' lives.

Improving lung cancer diagnoses: a machine learning approach for detection and prediction in CT-Scan image analysis

Improving lung cancer diagnoses: a machine learning approach for detection and prediction in CT-Scan image analysis

Pre-trained quantum convolutional neural network for COVID-19 disease classification using computed tomography images

Background The COVID-19 pandemic has had a significant influence on economies and healthcare systems around the globe. One of the most important strategies that has proven to be effective in limiting the disease and reducing its rapid spread is early detection and quick isolation of infections. Several diagnostic tools are currently being used for COVID-19 detection using computed tomography (CT) scan and chest X-ray (CXR) images. Methods In this study, a novel deep learning-based model is proposed for rapid detection of COVID-19 using CT-scan images. The model, called pre-trained quantum convolutional neural network (QCNN), seamlessly combines the strength of quantum computing with the feature extraction capabilities of a pre-trained convolutional neural network (CNN), particularly VGG16. By combining the robust feature learning of classical models with the complex data handling of quantum computing, the combination of QCNN and the pre-trained VGG16 model improves the accuracy of feature extraction and classification, which is the significance of the proposed model compared to classical and quantum-based models in previous works. Results The QCNN model was tested on a SARS-CoV-2 CT dataset, initially without any pre-trained models and then with a variety of pre-trained models, such as ResNet50, ResNet18, VGG16, VGG19, and EfficientNetV2L. The results showed the VGG16 model performs the best. The proposed model achieved 96.78% accuracy, 0.9837 precision, 0.9528 recall, 0.9835 specificity, 0.9678 F1-Score and 0.1373 loss. Conclusion Our study presents pre-trained QCNN models as a viable technique for COVID-19 disease detection, showcasing their effectiveness in reaching higher accuracy and specificity. The current paper adds to the continuous efforts to utilize artificial intelligence to aid healthcare professionals in the diagnosis of COVID-19 patients.

Correction: Chaotic Satin Bowerbird Optimizer Based Advanced AI Techniques for Detection of COVID‑19 Diseases from CT Scans Images

Correction: Chaotic Satin Bowerbird Optimizer Based Advanced AI Techniques for Detection of COVID‑19 Diseases from CT Scans Images

Measurement of Hydraulic Fracture Aperture by Electromagnetic Induction.

We present a new method for accurately measuring the aperture of a fluid-driven fracture. This method uses an eddy current probe located within a completion tool specifically designed to obtain the fracture aperture in the wellbore at the location where the fluid is injected into the fracture. The probe induces an eddy current in a target object, producing a magnetic field that affects the overall magnetic field. It does not have any limitations with respect to fluid pressure and temperature within a large range, making it unlike other methods. We demonstrate the accuracy and performance of the sensor under laboratory conditions. A hydraulic fracture experiment in a porous sandstone is conducted and discussed. The obtained measurement of the evolution of the fracture inlet aperture by the eddy current probe during the multiple injection cycles performed provided robust information. The residual fracture aperture (after the test) measured by the probe is in line with estimations from image processing of X-ray CT scan images as well as a thin-section analysis of sub-parts of the fractured specimen. The robustness and accuracy of this electromagnetic induction probe demonstrated herein under laboratory conditions indicate an interesting potential for field deployment.