Scanner Type Research Articles

Effect of intraoral scanner and fixed partial denture situation on the scan accuracy of multiple implants: An in vitro study.

Accuracy of intraoral implant scans may be affected by the region of the implant and the type of the intraoral scanner (IOSs). However, there is limited knowledge on the scan accuracy of multiple implants placed for an implant-supported fixed partial denture (FPD) in different partially edentulous situations when digitized by using different IOSs. To investigate the effect of IOS and FPD situation on the scan accuracy of two implants when partial-arch scans were performed. Tissue level implants were placed in 3 maxillary models with implant spaces either at right first premolar and right first molar sites (Model 1, 3-unit FPD), at right canine and right first molar sites (Model 2, 4-unit FPD), or at lateral incisor sites (Model 3, 4-unit FPD). Reference standard tessellation language (STL) files of the models were generated by using an optical scanner (ATOS Capsule 200MV120). Two IOSs (CEREC Primescan [CP] and TRIOS 3 [TR]) were used to perform partial-arch scans (test-scans) of each model (n=14), which were exported in STL format. A metrology-grade analysis software (GOM Inspect 2018) was used to superimpose test-scan STLs over the reference STL to calculate 3D distance, inter-implant distance, and angular (mesiodistal and buccopalatal) deviations. Trueness and precision analyses were performed by using bootstrap analysis of variance followed by Welch tests with Holm correction (α=0.05). Trueness of the scans was affected by IOS and FPD situation when 3D distance deviations were considered, while inter-implant distance, mesiodistal angular, and buccopalatal angular deviations were only affected by the FPD situation (p < 0.001). Scan precision was affected by the interaction between the IOSs and the FPD situation when 3D distance and buccopalatal angular deviations were concerned, while IOSs and FPD situation were effective when all deviations were concerned (p≤ 0.001). When 3D distance deviations were considered, CP scans had higher accuracy TR scans in Models 1 and 3 (p ≤ 0.002), and the Model 1 scans had the highest accuracy (p < 0.001). When inter-implant distance deviations were considered, Model 1 scans had the highest accuracy with CP and higher accuracy than Model 2 when TR was used (p ≤ 0.030). When mesiodistal angular deviations were considered, Model 1 scans had the highest accuracy (p ≤ 0.040). When buccopalatal angular deviations were considered, Model 1 scans had the highest accuracy among models when CP was used (p ≤ 0.020). Posterior 3-unit fixed partial denture implant scans, CP scans, and combination of these two factors had accuracy either similar to or better than their tested counterparts.

Preanalytic variable effects on segmentation and quantification machine learning algorithms for amyloid-β analyses on digitized human brain slides.

Computational machine learning (ML)-based frameworks could be advantageous for scalable analyses in neuropathology. A recent deep learning (DL) framework has shown promise in automating the processes of visualizing and quantifying different types of amyloid-β deposits as well as segmenting white matter (WM) from gray matter (GM) on digitized immunohistochemically stained slides. However, this framework has only been trained and evaluated on amyloid-β-stained slides with minimal changes in preanalytic variables. In this study, we evaluated select preanalytical variables including magnification, compression rate, and storage format using three digital slides scanners (Zeiss Axioscan Z1, Leica Aperio AT2, and Leica Aperio GT 450), on over 60 whole slide images, in a cohort of 14 cases having a spectrum of amyloid-β deposits. We conducted statistical comparisons of preanalytic variables with repeated measures analysis of variance evaluating the outputs of two DL frameworks for segmentation and object classification tasks. For both WM/GM segmentation and amyloid-β plaque classification tasks, there were statistical differences with respect to scanner types (p < 0.05) and magnifications (p < 0.05). Although small numbers of cases were analyzed, this pilot study highlights the significance of preanalytic variables that may alter the performance of ML algorithms.

Dedicated phantom tools using traceable 68Ge/68Ga point-like sources for dedicated-breast PET and positron emission mammography scanners.

Since the early 2000s, many types of positron emission tomography (PET) scanners dedicated to breast imaging for the diagnosis of breast cancer have been introduced. However, conventional performance evaluation methods developed for whole-body PET scanners cannot be used for such devices. In this study, we developed phantom tools for evaluating the quantitative accuracy of positron emission mammography (PEM) and dedicated-breast PET (dbPET) scanners using novel traceable point-like 68Ge/68Ga sources. The PEM phantom consisted of an acrylic cube (100 × 100 × 40mm) and three point-like sources. The dbPET phantom comprised an acrylic cylinder (ø100 × 100mm) and five point-like sources. These phantoms were used for evaluating the fundamental responses of clinical PEM and dbPET scanners to point-like inputs in a medium. The results showed that reasonable recovery values were obtained based on region-of-interest analyses of the reconstructed images. The developed phantoms using traceable 68Ge/68Ga point-like sources were useful for evaluating the physical characteristics of PEM and dbPET scanners. Thus, they offer a practical, reliable, and universal measurement scheme for evaluating various types of PET scanners using common sets of sealed sources.

An Overview on Image-Based and Scanner-Based 3D Modeling Technologies.

Advances in the scientific fields of photogrammetry and computer vision have led to the development of automated multi-image methods that solve the problem of 3D reconstruction. Simultaneously, 3D scanners have become a common source of data acquisition for 3D modeling of real objects/scenes/human bodies. This article presents a comprehensive overview of different 3D modeling technologies that may be used to generate 3D reconstructions of outer or inner surfaces of different kinds of targets. In this context, it covers the topics of 3D modeling using images via different methods, it provides a detailed classification of 3D scanners by additionally presenting the basic operating principles of each type of scanner, and it discusses the problem of generating 3D models from scans. Finally, it outlines some applications of 3D modeling, beyond well-established topographic ones.

Comparison of Diagnostic Quality of Multidetector Computed Tomography and Cone-Beam Computed Tomography in High and Low Resolution Modes for Assessment of the Nasal Cavity.

We aimed to compare the diagnostic quality of multidetector computed tomography (MDCT) and cone-beam computed tomography (CBCT) in high (HR) and low (LR) resolution modes for assessment of the nasal cavity and paranasal sinuses. This in vitro study was conducted on 5 dry human skulls by using a CBCT and a MDCT scanner in HR and LR modes to assess their diagnostic quality for 21 anatomical landmarks of the nose. The quality of images was evaluated by two oral and maxillofacial radiologists and a dentist using a four-point Likert scale of (I) poor, (II) decreased, (III) good, and (IV) excellent. Data were analyzed by STATA at 95% confidence interval. The Chi-square test was applied to compare the quality of visualization of landmarks based on the type of scanner. The diagnostic quality of HR CBCT and CT for the majority of landmarks was higher than that of LR CBCT and CT (P<0.05). The diagnostic quality of HR CBCT for agger nasi cells (P=0.010), olfactory cleft (P=0.032), sphenoethmoidal recess (P=0.032), and nasolacrimal duct (P=0.014) and LR CBCT for the middle turbinate (P=0.046) and middle meatus (P=0.031) was significantly higher than that of MDCT. The diagnostic quality of HR CBCT and CT for the majority of the landmarks in the nasal cavity and paranasal sinuses was higher than that of LR CBCT and CT. For the majority of landmarks, the diagnostic quality of CBCT and CT was the same; while for some landmarks, the diagnostic quality of HR and LR CBCT was higher than HR and LR CT. In general, CBCT has high efficacy for evaluation of the paranasal sinuses and the nasal cavity, and provides diagnostic information comparable to those provided by CT, but with a much lower radiation dose.

DeepEZ: A Graph Convolutional Network for Automated Epileptogenic Zone Localization From Resting-State fMRI Connectivity.

Epileptogenic zone (EZ) localization is a crucial step during diagnostic work up and therapeutic planning in medication refractory epilepsy. In this paper, we present the first deep learning approach to localize the EZ based on resting-state fMRI (rs-fMRI) data. Our network, called DeepEZ, uses a cascade of graph convolutions that emphasize signal propagation along expected anatomical pathways. We also integrate domain-specific information, such as an asymmetry term on the predicted EZ and a learned subject-specific bias to mitigate environmental confounds. We validate DeepEZ on rs-fMRI collected from 14 patients with focal epilepsy at the University of Wisconsin Madison. Using cross validation, we demonstrate that DeepEZ achieves consistently high EZ localization performance (Accuracy: 0.88 ± 0.03; AUC: 0.73 ± 0.03) that far outstripped any of the baseline methods. This performance is notable given the variability in EZ locations and scanner type across the cohort. Our results highlight the promise of using DeepEZ as an accurate and noninvasive therapeutic planning tool for medication refractory epilepsy. While prior work in EZ localization focused on identifying localized aberrant signatures, there is growing evidence that epileptic seizures affect inter-regional connectivity in the brain. DeepEZ allows clinicians to harness this information from noninvasive imaging that can easily be integrated into the existing clinical workflow.

Difficulties and Recommendations for AI-Based Prediction of Prostate Cancer Aggressiveness in Digital Pathology

Prostate cancer is among the most common cancers in men with around 1.4 million new cases each year world-wide. A vital part in the diagnosis of prostate cancer is the evaluation of its severity using biopsies and histopathology. Recent progress in artificial intelligence-based image analysis has led to a flurry of algorithms for the automated analysis of prostate cancer histopathological data focusing on the detection of cancerous areas, the grading of cancer severity, and patient outcome. Some of these approaches have reached human expert-level performance and digital models trained directly on patient outcomes might surpass human performance in the future. Although these results hold great promise for the future usage of digital pathology in clinical settings, several bottlenecks remain to be addressed. Especially the robustness, reliability and trustworthiness of predictions must be guaranteed across a wide range of variation in protocols and instrumentation. While human experts are relatively robust to technical and biological variation in biopsies, artificial intelligence-based systems tend to struggle with differences in staining intensity, color, scanner type, and image resolution, impeding the clinical usage of digital models. In this work we highlight salient problems and minimal requirements of computational pathology for future use in clinical settings, while focusing on prostate cancer as a use case. In particular, we highlight data and model problems and solutions that include data variability, dataset size, and data annotations, as well as model robustness to data heterogeneity, model prediction confidence, and the explainability of model decisions. While model and data requirements for successful computational pathology in clinics will be highlighted, legal, ethical, and deployment requirements will not be addressed in this review. In summary, we provide a short overview of the field, salient problems, and potential solutions to harvest the full potential of digital pathology for prostate cancer in clinical practice.

Distilling BlackBox to Interpretable Models for Efficient Transfer Learning.

Building generalizable AI models is one of the primary challenges in the healthcare domain. While radiologists rely on generalizable descriptive rules of abnormality, Neural Network (NN) models suffer even with a slight shift in input distribution (e.g., scanner type). Fine-tuning a model to transfer knowledge from one domain to another requires a significant amount of labeled data in the target domain. In this paper, we develop an interpretable model that can be efficiently fine-tuned to an unseen target domain with minimal computational cost. We assume the interpretable component of NN to be approximately domain-invariant. However, interpretable models typically underperform compared to their Blackbox (BB) variants. We start with a BB in the source domain and distill it into a mixture of shallow interpretable models using human-understandable concepts. As each interpretable model covers a subset of data, a mixture of interpretable models achieves comparable performance as BB. Further, we use the pseudo-labeling technique from semi-supervised learning (SSL) to learn the concept classifier in the target domain, followed by fine-tuning the interpretable models in the target domain. We evaluate our model using a real-life large-scale chest-X-ray (CXR) classification dataset. The code is available at: https://github.com/batmanlab/MICCAI-2023-Route-interpret-repeat-CXRs.

Метод оценки радиационного риска медицинского облучения при прохождении компьютерной томографии детей и подростков на основе данных протоколов сканирования

Computed tomography (CT) scan is a fairly accurate and reliable method for diagnosing various diseases, however, based on irradiation of the patient with ionizing radiation, it can lead to an excess risk of developing malignant neoplasms. According to the safety standards of the International Atomic Energy Agency, no patient can be exposed to medical radiation unless he or she has been informed of the risks associated with radiation exposure. This requirement is also contained in the Russian Radiation Safety Standards (NRB-99/2009). Radiation risk is estimated on the basis of calculated values of equivalent doses in the patient's organs and tissues using radiation risk models of the International Commission on Radiological Protection. This article is devoted to the improvement of the method for calculating equivalent doses in the organs and tissues of patients with computed tomography scan for further assessment of radiation risks. For 15 modern computed tomography scanners, with different geometry of patient scanning, the data necessary for calculating the equivalent doses in organs and tissues based on the parameters of the children's phantom were updated. The coefficients of conversion of the surrogate measure of the patient's radiation dose for the entire CT scan study, namely, the value of the "Dose Length Product" (DLP), into the values of equivalent doses in individual organs and tissues exposed to radiation were determined. The results of the calculation are given for five typical scan areas: chest, abdomen, pelvis, head and neck. In the course of simulation numerical modeling, organs and tissues with the highest equivalent doses were determined during CT scan examinations of children and adolescents. Unlike adult patients, for children and adolescents, the dose load on the surface of the bone is significantly lower and does not stand out in any of the considered types of CT scan examinations. It should be noted that with CT scan of the cervical spine, due to the specifics of the geometry of scanning new types of tomography scanners, the dose load on the thyroid gland of boys increased. The dose conversion factors obtained in this paper with estimates of standard deviations make it possible to assess the radiation risks of children and adolescents during examinations on modern CT scanners and optimize their dose loads.

СУЧАСНІ 3D-ТЕХНОЛОГІЇ В МАШИНОБУДУВАННІ ТА АВТОМОБІЛЬНОМУ ТРАНСПОРТІ

The trends in the development of manufacturing industries are aimed at highly efficient design. 3D technologies play the main role in product design. Therefore, a mandatory condition for the high-quality training of specialists in the fields of knowledge "Road transport" and "Industrial engineering" is the study of the current state and trends in the development of 3D technologies.The rapid development of new technologies for the production of parts, in particular additive ones, makes it necessary to change the approach to the training of highly qualified workers. The approach to the presentation of information should be comprehensive, highlighting all the interrelationships between the stages of design and production. A significant number of works are devoted to spatial modeling in SolidWorks, Delcam PowerShape, Сatia, etc. The main principles of creating models are described, the advantages of the program and the scope of application are defined. 3D print-ing and scanning technologies are developing rapidly.However, there is currently no comprehensive methodological approach to the study of modern 3D technologies, when training specialists in technical areas.The main goal is to study the state of development and the field of use of modern 3D technologies in automobile transport and industrial engineering.The types and scope of use of 3D printers and scanners are analyzed. Positive and negative features of representatives of each type are determined.The main aspects of teaching the discipline "Modern 3D technologies" are considered. The methods of creating spatial models in the SolidWorks system are considered. The most advanced 3D printing technologies and materials used are de-scribed. The principle of operation and types of 3D scanners are reviewed.

Методика оценки радиационного риска медицинского облучения при прохождении компьютерной томографии с учётом неопределённостей модели риска

The paper presents analysis of different factors affecting the uncertainty model for estimating ra-diation risk from computed tomography (CT). Uncertainties in radiation doses estimates caused by measurement (instrumental) errors or used dose estimation methods, the size of the scanned area and the type of CT scanner. The uncertainty of radiation dose due to measurement errors or dose estimation methods, the size of the scanned area and type of a CT scan may cause the uncertainties as well. Data used for calculating equivalent doses in individual organs and tissues and for calculating lifetime radiation risk of cancer development due to routine CT testing were updated. Conversion factors for DLP, a measure of radiation dose a patient received during CT exams of thoracic, abdomen and head organs, were determined and used for conversion of the CT doses to equivalent doses for individual organs and tissues exposed to radiation. Data for 15 state-of-the-art CT scanners with varying scanning geometry were updated. Uncertainties in life-time radiation risk were determined by estimating 95% confidence intervals for mean dose-proportionality ratios. Standard deviations related to specific dose distribution, scanning geometry and other factors that impact on uncertainty of radiation risk estimates were calculated. The standard deviations associated with the specifics of the dose distribution, scanning geometry and other factors affecting the uncertainties of radiation risk assessments were calculated. In the course of simulation modelling, organs and tissues were identified that are most exposed to radiation during CT of the chest, abdomen and head.

Lesion Detection Through MRI Postprocessing in Pathology-Proven Focal Cortical Dysplasia: Experience at a Single Institution in the Republic of Korea.

Focal cortical dysplasia (FCD) is one of the most common causes of drug-resistant epilepsy, and necessitates a multimodal evaluation to ensure optimal surgical treatment. This study aimed to determine the supportive value of the morphometric analysis program (MAP) in detecting FCD using data from a single institution in Korea. To develop a standard reference for the MAP, normal-looking MRIs by two scanners that are frequently used in this center were chosen. Patients with drug-resistant epilepsy and FCD after surgery were candidates for the analysis. The three-dimensional T1-weighted MRI scans of the patients were analyzed as test cases using the MAP. The MRI scans of 87 patients were included in the analysis. The radiologist detected abnormal findings correlated with FCD (RAD positive [RAD(+)]) in 34 cases (39.1%), while the MAP could detect FCD in 25.3% of cases. A combination of the MAP (MAP[+] cases) with interpretations by the radiologist increased the detection to 42.5% (37 cases). The lesion detection rate was not different according to the type of reference scanners except in one case. MAP(+)/RAD(-) presented in three cases, all of which had FCD type IIa. The detection rate was slightly higher using the same kind of scanner as a reference, but not significantly (35.0% vs. 22.4% p=0.26). The results of postprocessing in the MAP for detecting FCD did not depend on the type of reference scanner, and the MAP was the strongest in detecting FCD IIa. We suggested that the MAP could be widely utilized without developing institutional standards and could become an effective tool for detecting FCD lesions.

The Application, Accuracy, and Utilization of Computer-Aided Design/Computer-Aided Manufacturing (CAD/CAM) Systems in Restorative Dentistry

Restorative dentistry is crucial for improving oral health and restoring damaged teeth. Computer-Aided Design/Computer-Aided Manufacturing (CAD/CAM) systems have revolutionized the field by integrating digital technologies, offering benefits over traditional methods. These systems streamline the restorative workflow, provide accurate restorations with excellent fit and aesthetics, and reduce chairside and laboratory processing times. CAD/CAM applications range from single-tooth restorations to fixed dental prostheses, implant-supported restorations, and orthodontic appliances. The accuracy and precision of CAD/CAM systems are evaluated through various methods, and factors influencing accuracy include scanner type, software algorithms, milling/3D printing technology, and materials used. Strategies to improve accuracy include optimizing scanning techniques and using advanced software algorithms. CAD/CAM systems have significant clinical implications, and their future holds promising advancements in intraoral scanning, CAD software, and CAM technologies. However, challenges such as initial investment, operator skill, and material selection need to be considered.

Performance of a deep convolutional neural network for MRI-based vertebral body measurements and insufficiency fracture detection

ObjectivesThe aim is to validate the performance of a deep convolutional neural network (DCNN) for vertebral body measurements and insufficiency fracture detection on lumbar spine MRI.MethodsThis retrospective analysis included 1000 vertebral bodies in 200 patients (age 75.2 ± 9.8 years) who underwent lumbar spine MRI at multiple institutions. 160/200 patients had ≥ one vertebral body insufficiency fracture, 40/200 had no fracture. The performance of the DCNN and that of two fellowship-trained musculoskeletal radiologists in vertebral body measurements (anterior/posterior height, extent of endplate concavity, vertebral angle) and evaluation for insufficiency fractures were compared. Statistics included (a) interobserver reliability metrics using intraclass correlation coefficient (ICC), kappa statistics, and Bland-Altman analysis, and (b) diagnostic performance metrics (sensitivity, specificity, accuracy). A statistically significant difference was accepted if the 95% confidence intervals did not overlap.ResultsThe inter-reader agreement between radiologists and the DCNN was excellent for vertebral body measurements, with ICC values of > 0.94 for anterior and posterior vertebral height and vertebral angle, and good to excellent for superior and inferior endplate concavity with ICC values of 0.79–0.85. The performance of the DCNN in fracture detection yielded a sensitivity of 0.941 (0.903–0.968), specificity of 0.969 (0.954–0.980), and accuracy of 0.962 (0.948–0.973). The diagnostic performance of the DCNN was independent of the radiological institution (accuracy 0.964 vs. 0.960), type of MRI scanner (accuracy 0.957 vs. 0.964), and magnetic field strength (accuracy 0.966 vs. 0.957).ConclusionsA DCNN can achieve high diagnostic performance in vertebral body measurements and insufficiency fracture detection on heterogeneous lumbar spine MRI.Key Points• A DCNN has the potential for high diagnostic performance in measuring vertebral bodies and detecting insufficiency fractures of the lumbar spine.

Effects of Processing Delays in Different Coverages on the Image Quality of Photostimulable Phosphor Plates With Acteon and Digora Scanners

Background: Delay in the processing of photostimulable phosphor storage plates (PSP) is a common occurrence in crowded clinics. Accordingly, the effects of processing delays in different coverages on the image quality of photostimulable PSPs were investigated with Acteon and Digora scanners. Methods: Three Acteon (group A) and three Digora (group B) PSPs were used in this in vitro study. Each group had three subgroups according to three coverages, including protective box (A1 , B1 ), semitransparent (A2 , B2 ), and original dark case plates (A3 , B3 ). An aluminum step wedge was subjected to constant exposure conditions. The exposed plates were immediately processed with their corresponding scanner device (the golden standard), 5, 10, 20, 30, 40, and 60 minutes after exposure. The average gray level information of the 2nd, 5th, and 8th steps of the Al wedge was considered as the mean gray values (MGVs) of each wedge. The difference between the gray values of the 8th and 2nd steps was measured as image contrast. Results: There was a significant difference between the contrast and MGVs of Acteon and Digora PSPs at all processing delay times (P&lt;0.05). In general, there was no significant difference in the image MGVs and contrast between subgroups in any of the scanners (total P&gt;0.05). In each subgroup, MGVs increased, while contrast decreased by increasing the processing delay time; the difference was significant except for the MGVs in the first 5 minutes of A1 (P=0.12) and A3 (P=0.06). Conclusions: Thus, the type of scanner was effective on image quality; the type of PSP coating in the first few minutes could affect the rate of image quality loss. However, the scan time had a greater effect on the amount of image loss.

A new approach to dose reference levels in pediatric CT: Age and size-specific dose estimation

BackgroundAlthough paediatrics' are more radio-sensitivity than adults, the number of requested paediatric CT examinations is increasing globally. Significant efforts have been focused on achieving the lowest possible radiation dose for paediatric patients, particularly adjusting for size differences. This study aims to establish Diagnostic Reference Levels (DRLs) for paediatric patients based on size-specific dose estimates (SSDE). MethodIn this retrospective national study, CT scans for brain, chest, abdominopelvic, and chest, abdomen, and pelvis (CAP) were collected from four hospitals in Jordan undergoing paediatric CT scans. A total of 1818 cases were randomly selected in four age categories (<1 year, 1–4 years, 5–10 years, 11–18 years). The SSDE values were determined by multiplying the volume CT dose index (CTDIvol) with the conversion factor extracted from the American Association of Physicists in Medicine Report 204. ResultsVariations exist between the DRLs between the different hospitals, age groups, and variations in protocols with different types of CT scanners. The DRL values (CTDIvol, dose-length product (DLP) and SSDE) for the four age categories were as follows; <1 year: brain (47.88 mGy, 741.67 mGy cm and 58.40 mGy), chest (5.65 mGy, 124 mGy cm and 13.91 mGy), abdominopelvic (12.65 mGy, 321.5 mGy cm and 28.72 mGy) and CAP (16.12 mGy, 507.72 mGy cm and 38.04 mGy). 1–4 years, brain (54.79 mGy, 979.12 mGy cm and 55.88 mGy), chest (7.37 mGy, 220.85 mGy cm and 14.68 mGy), abdominopelvic (16.16 mGy, 424.72 mGy cm and 32.68 mGy) and CAP (16.13 mGy, 742.1 mGy cm and 33.54 mGy). 5–10 years: brain (65.03 mGy, 1129.94 mGy cm and 55.92 mGy), chest (12.57 mGy, 383.9 mGy cm and 22.45 mGy), abdominopelvic (12.34 mGy, 450.75 mGy cm and 22.23 mGy) and CAP (13.46 mGy, 748.85 mGy cm and 25.69 mGy). 11–18 years, brain (60.7 mGy, 1207.9 mGy cm and 41.81 mGy), chest examination (12.94 mGy, 496.2 mGy cm and 20.49 mGy), abdominopelvic (16.13 mGy, 803.07 mGy cm and 23.06 mGy) and CAP (16.13 mGy, 1101.5 mGy cm and 23.85 mGy). ConclusionThere were increases in CTDIvol, DLP, and SSDE with ascending age groups. SSDE and age are closely matched to delivered radiation in paediatric CT; however, radiation dose levels remain high in Jordan. This work highlights the need for caution when administering radiation in the paediatric population.

Robust and accurate pulmonary nodule detection with self-supervised feature learning on domain adaptation.

Medical imaging data annotation is expensive and time-consuming. Supervised deep learning approaches may encounter overfitting if trained with limited medical data, and further affect the robustness of computer-aided diagnosis (CAD) on CT scans collected by various scanner vendors. Additionally, the high false-positive rate in automatic lung nodule detection methods prevents their applications in daily clinical routine diagnosis. To tackle these issues, we first introduce a novel self-learning schema to train a pre-trained model by learning rich feature representatives from large-scale unlabeled data without extra annotation, which guarantees a consistent detection performance over novel datasets. Then, a 3D feature pyramid network (3DFPN) is proposed for high-sensitivity nodule detection by extracting multi-scale features, where the weights of the backbone network are initialized by the pre-trained model and then fine-tuned in a supervised manner. Further, a High Sensitivity and Specificity (HS) network is proposed to reduce false positives by tracking the appearance changes among continuous CT slices on Location History Images (LHI) for the detected nodule candidates. The proposed method's performance and robustness are evaluated on several publicly available datasets, including LUNA16, SPIE-AAPM, LungTIME, and HMS. Our proposed detector achieves the state-of-the-art result of sensitivity at false positive per scan on the LUNA16 dataset. The proposed framework's generalizability has been evaluated on three additional datasets (i.e., SPIE-AAPM, LungTIME, and HMS) captured by different types of CT scanners.

The Determination of Diagnostic Reference Range (DRR) Based on Clinical Indications in Head Computed Tomography (CT) Imaging

Diagnostic reference range (DRR) were introduced to help manage image quality as well as patient dose by providing minimum estimated dose that covers 25–75% range of dose distribution. The International Commission on Radiological Protection (ICRP) (2017) recommended 75th percentile as upper boundary and 25th percentile as lower boundary of DRR [1]. Dose below the lower range should compromised accurate interpretation of the image and dose above the upper range may be in excess and need to be reviewed. Meanwhile, Diagnostic Reference Levels (DRLs) also recommended by ICRP as a guidance to identify any uncommon high dose delivered by computed tomography (CT) scanner. DRLs are established at 75th percentile of dose distribution and should be implemented at international, regional, or local level for dose management.&#x0D; &#x0D; This study aims to establish a local DRR based on five most common clinical indications for adult head CT scanning at Department of Radiology, Hospital Universiti Sains Malaysia (HUSM), Kelantan. This study involved a retrospective survey on adult patient data undergone head CT scan with two CT scanners, Toshiba CT scanner at Trauma Department and Siemens CT scanner at Radiology Department, HUSM. The related data regarding the scanning protocols and dose descriptor (dose length product, DLP) were recorded for the most common clinical indication in head CT scans.&#x0D; &#x0D; Results showed the five most common clinical indication related to head CT scans were abscess, bleeding, stroke, trauma, and tumour. The DLP values were varied based on their clinical indications and CT scanner types. Figure 1 shows boxplots representing the DRRs of dose length product (DLP) based on the five most common clinical indications in head CT scans for Toshiba and Siemens CT scanners. From the figure, the DLP ranges for Toshiba CT scanner were slightly higher than Siemens CT scanners. The lowest DLP value was observed for tumour indication and the highest DLP was bleeding and trauma indications. Clinical indication that requires higher image resolution and details and longer scan length, recorded higher DLP distribution.&#x0D; This study demonstrates that the dose descriptor values, CTDIvol and DLP were varied for similar anatomical region based on different image quality requirement for each clinical indication. Therefore, DRR and DRL should be established based on specific clinical indication and routine dose audit is essential to ensure patients safety.

Diagnostic Reference Range (DRR) in Computed Tomography (CT) Imaging Based on Clinical Indications Related to Thorax, Abdomen and Pelvic Regions

The concept of diagnostic reference range (DRR) is introduced to address the balance between a patient’s risk (radiation dose) and benefit (image quality necessary for an accurate diagnosis). R provides a minimum estimated dose that covers 25–75% range of dose distribution. The International Commission on Radiological Protection, ICRP (2017) recommended the 75th percentile as the upper d 25th percentile as the lower limit of DRR [1]. Dose below the lower range should compromise accurate interpretation of the image and dose above the upper range may be in excess and need to be reviewed. Meanwhile, Diagnostic Reference Levels (DRLs) are set at the 75th percentile of dose distribution and recommended by ICRP as a guidance to identify any unusual computed tomography (CT) procedures that consistently exceed the established DRLs.&#x0D; The aim of this study was to establish the local DRR for CT scanning based on clinical indications associated to thorax, abdomen, and pelvic regions in adult patients at Department of Radiology, Hospital Universiti Sains Malaysia (HUSM), Kelantan. This study involved a retrospective survey on data of adult patients underwent CT scanning based on 5 most common clinical indications related to thorax, abdomen, and pelvic regions at two departments, Radiology and Trauma Department, HUSM. This study obtained human ethical approval from the university ethical committee (Protocol code: USM/JEPeM/22010025). The CT dose metrics (dose length product, DLP and total DLP) and associated technical factors were collected.&#x0D; Results showed that the five most common clinical indications were cancer staging, coronary artery disease, pulmonary embolism, stones (CTU protocol) and stones (KUB protocol). The DLP values were varied based on their clinical indications and CT scanner types. Figure 1 shows boxplots representing the DRRs of DLP based on five most common clinical indications associated to CT scans of thorax, abdomen, and pelvic regions for Toshiba and Siemens CT scanners. The highest total DLP values was 4722 mGy.cm observed from CT scanning of cancer staging using Siemens scanner. While the lowest DLP value is 40 mGy.cm for stones scanning using the KUB protocol also using Siemens scanner. The main factor that contributes to higher DLP is the longer scanning length.&#x0D; This study demonstrates that the dose descriptor values, DLP and total DLP were varied for similar anatomical region based on different image quality requirement for each clinical indication. Thus, utilisation of both DRR and DRL are crucial to minimise the unnecessary dose to patient and ensure the optimum image quality is produced by each CT scanning.

Performance variability of radiomics machine learning models for the detection of clinically significant prostate cancer in heterogeneous MRI datasets.

Radiomics promises to enhance the discriminative performance for clinically significant prostate cancer (csPCa), but still lacks validation in real-life scenarios. This study investigates the classification performance and robustness of machine learning radiomics models in heterogeneous MRI datasets to characterize suspicious prostate lesions for non-invasive prediction of prostate cancer (PCa) aggressiveness compared to conventional imaging biomarkers. A total of 142 patients with clinical suspicion of PCa underwent 1.5T or 3T biparametric MRI (7 scanner types, 14 institutions) and exhibited suspicious lesions [prostate Imaging Reporting and Data System (PI-RADS) score ≥3] in peripheral or transitional zones. Whole-gland and index-lesion segmentations were performed semi-automatically. A total of 1,482 quantitative morphologic, shape, texture, and intensity-based radiomics features were extracted from T2-weighted and apparent diffusion coefficient (ADC)-images and assessed using random forest and logistic regression models. Five-fold cross-validation performance in terms of area under the ROC curve was compared to mean ADC (mADC), PI-RADS and prostate-specific antigen density (PSAD). Bias mitigation techniques targeting the high-dimensional feature space and inherent class imbalance were applied and robustness of results was systematically evaluated. Trained models showed mean area under the curves (AUCs) ranging from 0.78 to 0.83 in csPCa classification. Despite using mitigation techniques, high performance variability of results could be demonstrated. Trained models achieved on average numerically higher classification performance compared to clinical parameters PI-RADS (AUC =0.78), mADC (AUC =0.71) and PSAD (AUC =0.63). Radiomics models' classification performance of csPCa was numerically but not significantly higher than PI-RADS scoring. Overall, clinical applicability in heterogeneous MRI datasets is limited because of high variability of results. Performance variability, robustness and reproducibility of radiomics-based measures should be addressed more transparently in future research to enable broad clinical application.