Enhancing Reproductive Organ Segmentation in Pediatric CT via Adversarial Learning.

Organ segmentation from CT images is critical in the early diagnosis of diseases, progress monitoring, pre-operative planning, radiation therapy planning, and CT dose estimation. However, data limitation remains one of the main challenges in medical image segmentation tasks. This challenge is particularly huge in pediatric CT segmentation due to children's heightened sensitivity to radiation. In order to address this issue, we propose a novel segmentation framework with a built-in auxiliary classifier generative adversarial network (ACGAN) that conditions age, simultaneously generating additional features during training. The proposed conditional feature generation segmentation network (CFG-SegNet) was trained on a single loss function and used 2.5D segmentation batches. Our experiment was performed on a dataset with 359 subjects (180 male and 179 female) aged from 5 days to 16 years and a mean age of 7 years. CFG-SegNet achieved an average segmentation accuracy of 0.681 dice similarity coefficient (DSC) on the prostate, 0.619 DSC on the uterus, 0.912 DSC on the liver, and 0.832 DSC on the heart with four-fold cross-validation. We compared the segmentation accuracy of our proposed method with previously published U-Net results, and our network improved the segmentation accuracy by 2.7%, 2.6%, 2.8%, and 3.4% for the prostate, uterus, liver, and heart, respectively. The results indicate that our high-performing segmentation framework can more precisely segment organs when limited training images are available.

Deep learning is a popular and powerful tool in computed tomography (CT) image processing such as organ segmentation, but its requirement of large training datasets remains a challenge. Even though there is a large anatomical variability for children during their growth, the training datasets for pediatric CT scans are especially hard to obtain due to risks of radiation to children. In this paper, we propose a method to conditionally synthesize realistic pediatric CT images using a new auxiliary classifier generative adversarial network (ACGAN) architecture by taking age information into account. The proposed network generated age-conditioned high-resolution CT images to enrich pediatric training datasets.

Practical Strategies to Reduce Pediatric CT Radiation Dose

Automatic segmentation of temporal bone computed tomography (CT) images is fundamental to image-guided otologic surgery and the intelligent analysis of CT images in the field of otology. This study was conducted to test a convolutional neural network (CNN) model that can automatically segment almost all temporal bone anatomy structures in adult and pediatric CT images. A dataset comprising 80 annotated CT volumes was collected, of which 40 samples were obtained from adults and 40 from children. A further 60 annotated CT volumes (30 from adults and 30 from children) were used to train the model. The remaining 20 annotated CT volumes were employed to determine the model's generalizability for automatic segmentation. Finally, the Dice coefficient (DC) and average symmetric surface distance (ASSD) were utilized as metrics to evaluate the performance of the CNN model. Two independent-sample t-tests were used to compare the test set results of adults and children. In the adult test set, the mean DC values of all the structures ranged from 0.714 to 0.912, and the ASSD values were less than 0.24 mm for 11 structures. In the pediatric test set, the mean DC values of all the structures ranged from 0.658 to 0.915, and the ASSD values were less than 0.18 mm for 11 structures. There was no statistically significant difference between the adult and child test sets in most temporal bone structures. Our CNN model shows excellent automatic segmentation performance and good generalizability for both adult and pediatric temporal bone CT images, which can help to advance otologist education, intelligent imaging diagnosis, surgery simulation, application of augmented reality, and preoperative planning for image-guided otology surgery.

Reducing radiation to children: the resident’s role

Background: Children have a potential risk from radiation exposure because they are more sensitive to radiation than adults. Objective: The purpose of this work is to estimate image quality according to tube voltage (kV) and radiation dose in pediatric computed tomography (CT) using deep learning reconstruction (DLR). Methods: Phantom images of children and adults were obtained for kV, radiation dose, and image reconstruction methods. The CT emits a fan beam to the opposite detector, and the geometry of the detector was symmetrical. Phantom images of children and adults were acquired at a volume CT dose index (CTDIvol) from 0.5 to 10.0 mGy for tube voltages at 80, 100, and 120 kV. A DLR was used to reconstruct the phantom image, and filtered back projection (FBP) and iterative reconstruction (IR) were also performed for comparison with the DLR. Image quality was evaluated by measuring the contrast-to-noise ratio (CNR) and noise. Results: Under the same imaging conditions, the DLR images of pediatric and adult phantoms generally provided improved CNR and noise compared with the FBP and IR images. At a similar CNR and noise, the FBP, IR, and DLR of the pediatric images showed a dose reduction compared with the FBP, IR, and DLR of the adult images, respectively. In terms of the effect of tube voltage, the CNR of the 100 kV DLR images was higher than that of the 120 kV DLR images. Conclusion: According to the results, since pediatric CT images maintain the same image quality at lower doses compared with adult CT images, DLR can improve image quality while reducing the radiation dose in children’s abdominal CT scans.

Although advanced statistical iterative reconstruction (IR) techniques are valued in pediatric computed tomography (CT) imaging, there is little published data on how these techniques affect image quality and radiation dose in the pediatric population. This is particularly true in the context of pediatric head CT examinations. This study analyzed the differences in image quality and several standard metrics of radiation dose on multidetector pediatric head CT examinations performed using standard filtered back projection (FBP) with reconstructions using iDose, a fourth-generation statistical iterative reconstruction technique. Using a retrospective review of 282 pediatric head CT examinations, we compared how iDose fared against FBP for effects on several standard metrics of radiation dose and qualitative and quantitative assessment of image quality. Our assessment revealed that examinations obtained using low-dose protocols reconstructed using iDose, when compared with standard-dose examinations reconstructed using FBP, resulted in significant radiation dose reduction while performing equally or better in quantitative image quality parameters. For most qualitative image quality parameters, the iDose group demonstrated equal performance to standard filtered back technique with a few notable exceptions. In the parameter of image sharpness in the 1.5 to 7 year olds, iDose fared better than FBP. However, FBP outperformed iDose in the qualitative parameters of decreased image graininess/noise in patients older than 13 years, improved image sharpness in patients aged between 7 and 13 years, and improved visibility of small parts for those aged 7 to 13 years. We conclude that iDose is effective at allowing significant radiation dose reduction while maintaining or, rarely, even improving quantitative image quality compared with FBP in the setting of pediatric head CT examinations. However, for certain qualitative image quality parameters in older-aged children, the use of iDose resulted in a poorer performance compared with FBP.

Computed Tomography (CT) contributes significantly to medical radiation exposure, demanding rigorous dose optimization, especially for the radiosensitive pediatric population, yet high dose variability often results from inappropriate protocol selection. To address this, a custom-developed pediatric chest phantom, radiologically representative of a 7-year-old child, was utilized to systematically quantify dose impacts. The phantom was scanned using a Toshiba Aquilion Prime CT scanner across two protocol presets (adult chest and pediatric chest) at three tube voltage settings (80, 100, and 120 kV), with the pediatric protocol engaging the Automated Exposure Control (AEC). Dose metrics, including Volume CT Dose Index (CTDIvol) and Size-Specific Dose Estimate (SSDE), were calculated via the validated IndoseCT platform. The results demonstrated that the adult protocol consistently yielded SSDE values that were approximately 3.0 to 3.4 times greater than the pediatric protocol across all kV levels. Specifically, raising the tube voltage in the adult protocol caused a sharp rise in SSDE (from 2.10 mGy at 80 kV to 5.81 mGy at 120 kV). In contrast, the pediatric protocol maintained a highly stable SSDE (ranging from 1.52 mGy to 1.73 mGy) due to effective AEC compensation. In conclusion, the use of inappropriate adult presets for pediatric CT introduces a substantial and avoidable radiation dose penalty, potentially leading to doses that exceed established Diagnostic Reference Levels (DRLs). This study reinforces the critical necessity of strict adherence to size- and age-specific protocols and optimal AEC engagement as the primary strategy for achieving significant dose reduction in pediatric CT imaging. Keywords: Pediatric CT, Dose Reduction, SSDE, Tube Voltage, Protocol Optimization, Pediatric Phantom

Radiation dose for common pediatric computed tomography examinations and evaluation of clinical image quality in pediatric chest CT scans.

Clinical utility of measuring temporal muscle volume by head computed tomography for Global Leadership Initiative on Malnutrition phenotypic criteria in critically ill patients.

Background.Patients, especially children, are exposed to substantially high doses of ionising radiation during computed tomography (CT) procedures. Children are several times more susceptible to ionising radiation than adults. Diagnostic reference levels (DRLs) are an important tool for monitoring and optimising patient radiation exposure from radiological procedures. The aim of this study is to estimate the ionising radiation exposure doses and set local DRLs for head CT examinations according to age and to compare local DRLs with national and European DRLs and with literature data in other countries.Materials and methods.Scan parameters of single-phase head CT examinations were collected. Patients were grouped by age in the following intervals: <1, 1−5, 5−10, 10−15 and 15−18 years. Local age-based DRLs set as the 3rd quartile of the median dose-length product (DLP) were calculated. Literature analysis was performed on PubMed search engine on inclusion criteria: publication date 2015–2020, used keywords paediatric computed tomography, paediatric CT, diagnostic reference levels (DRLs). The 23 articles discussing paediatric DRLs were further analysed.Results.Data was collected from 194 paediatric head CT examinations performed in 2019. The median DLP values for head CT were 144.3, 233.7, 246.4, 288.9, 315.5 for <1, 1−5, 5−10, 10−15 and 15−18 years old groups. Estimated local DRLs for head CT examinations are 170, 300, 310, 320, 360 mGy*cm for <1, 1−5, 5−10, 10−15 and 15−18 years age groups respectively and 130, 210, 275, 320 mGy*cm for 0−3 months, 3 months−1 year, 1−6 years and ≥ 6 years age groups respectively. Conclusions.Results of this study showed that settled new local DRLs of head CT examinations were 2–4 times lower than national DRLs and about 2 times lower than European DRLs. Moreover, the study indicated that paediatric head CT doses are significantly lower in comparison with those indicated in the majority of published data from other hospitals over the last 6 years. Patient dose assessment and local DRLs establishment plays important role in future exposure optimisation.

The effective dose (ED) in computed tomography (CT) may be calculated by multiplying the dose-length product (DLP) by a conversion factor. As children grow, automatic exposure control increases the DLP, while the conversion factor decreases; these two changes affect the ED in opposite ways. The aim of this study was to investigate the methods of ED estimation according to age in pediatric brain CT. We retrospectively analyzed 980 brain CT scans performed for various clinical indications in children. The conversion factor at each age, in integer years, was determined based on the values at 0, 1, 5, and 10 years provided by the International Commission on Radiological Protection (ICRP), using a curve (curve method) or lines (linear method). In the simple method, the ED was estimated using the ICRP conversion factor for the closest age. We also analyzed the ED estimated by a radiation dose management system. Although the median DLP at each age increased with age, the median ED estimated by the curve method was highest at 0 years, decreased with age, and then plateaued at 9 years. The linear method yielded mildly different results, especially at 2 and 3 years. The ED estimated by the simple method or the radiation dose management system showed inconsistent, up-and-down changes with age. In conclusion, the ED in pediatric brain CT decreases with age despite increased DLP. Determination of the conversion factor at each age using a curve is expected to contribute to estimating the ED in pediatric CT according to age.

It is important to employ radiation dose reduction techniques in pediatric computed tomography (CT) to reduce potential risks of radiation‐induced malignancy. Automatic tube potential (kV) selection tools have been developed and become available on many CT scanners, which select the optimum kV based on the patient size and clinical task to improve the radiation dose efficiency. However, its use in pediatric CT has been mostly empirical, following manufacturer’s default recommendation without solid demonstration for quality improvement. This study aimed to implement an automatic tube potential tool (CAREkV, Siemens Healthcare) into routine pediatric CT practice, using the “Plan‐Do‐Study‐Act” quality improvement process, in place of an existing kV/mAs technique chart. The design of this quality improvement project involved Plan‐Do‐Study‐Act stages. Plan and Do stages identified the criteria for optimal automatic kV selection; a range of phantoms representing typical pediatric groups were scanned on a dual‐source 128‐slice scanner using a fast‐pitch scanning mode. The identified CAREkV settings were implemented into the CT protocol and evaluated after a 6‐month period. In the Study stage, an objective evaluation of the image metrics and radiation dose for two similar patient cohorts using CAREkV and the technique‐chart, respectively, were compared. The kV selected, image quality and radiation dose determined by CAREkV were comparable to those obtained while using the technique‐chart. The CAREkV was successfully implemented into our pediatric abdominopelvic CT practice. By utilizing the “PDSA” process optimal image quality and radiation dose reduction were achieved with an automatic kV selection tool to improve CT workflow.

Despite the fact that doses to paediatric patients from computed tomography (CT) examinations are of special concern, only few data or studies for setting of paediatric diagnostic reference levels (DRLs) have been published. In this study, doses to children were estimated from chest and head CT, in order to study the feasibility of DRLs for these examinations. It is shown that for the DRLs, patient dose data from different CT scanners should be collected in age or weight groups, possibly for different indications. For practical reasons, the DRLs for paediatric chest CT should be given as a continuous DRL curve as a function of patient weight. For paediatric head CT, DRLs for a few age groups could be given. The users of the DRLs should be aware of the calibration phantom applied in the console calibration for different paediatric scanning protocols. The feasibility of DRLs should be re-evaluated every 2-3 y.

Radiology| October 01 2002 Think Twice Before Ordering That CT AAP Grand Rounds (2002) 8 (4): 41–42. https://doi.org/10.1542/gr.8-4-41 Views Icon Views Article contents Figures & tables Video Audio Supplementary Data Peer Review Share Icon Share Twitter LinkedIn Tools Icon Tools Get Permissions Cite Icon Cite Search Site Citation Think Twice Before Ordering That CT. AAP Grand Rounds October 2002; 8 (4): 41–42. https://doi.org/10.1542/gr.8-4-41 Download citation file: Ris (Zotero) Reference Manager EasyBib Bookends Mendeley Papers EndNote RefWorks BibTex toolbar search nav search search input Search input auto suggest search filter All PublicationsAll JournalsAAP Grand RoundsPediatricsHospital PediatricsPediatrics In ReviewNeoReviewsAAP NewsAll AAP Sites Search Advanced Search Topics: cancer, computed tomography, diagnostic radiologic examination, radiation exposure, radiologists, radiotherapy dosage, cancer risk, spiral computed tomography Source: Society of Pediatric Radiology. The ALARA (As Low As Reasonably Achievable) concept in pediatric CT intelligent dose reduction. Multidisciplinary conference organized by the Society of Pediatric Radiology. August 18–19, 2001. Pediatr Radiol. 2002;32:217–313. On January 22, 2001, USA Today published a front page article entitled “Radiation from CT scans in children linked to cancer.”1 This controversial and somewhat distorted report sensationalized the results of 3 articles and an editorial published in the February 2001 issue of the American Journal of Roentgenology. Roentgenology 2–,4 The 3 articles included an estimation of cancer risk from pediatric computed tomography (CT), a critical evaluation of why the helical CT technical settings used throughout the general radiology community were not adjusted for children, and strategies on how to minimize radiation dose during pediatric body CT using a single detector helical scanner. A flood of public outcry hit the medical community, focused primarily on pediatricians. Concerned, fearful, and angry parents worried about their child’s CT examination, and in some cases blamed CT radiation for their child’s cancer. In response, the Society for Pediatric Radiology organized a conference in August 2001 to address the scientific aspects of this controversy and expand on the “ALARA” (As Low As Reasonably Achievable) concept that is generally applied by radiologists to any medically necessary radiation exposure. The following summarizes excerpts from those presentations. The use of pediatric CT exams has risen from 4% of all CT scans performed in 1989 to around 10% today. An estimated 2.7 million children receive CT exams every year in the United States. CT scans account for only 10% of x-ray exams but contribute a disproportionate 67% of the overall collective radiation dose. Multiple exams are often performed in the same patient with approximately 30% of patients having at least 3 CT exams; 11% have more. Children are more sensitive to radiation than adults by a factor of 10; girls are at more risk than boys by a factor of 2; and exposure at a younger age (especially below 15 years) carries a significantly higher risk of cancer. Since radiation-induced cancers appear at about the same age as spontaneous cancers of a similar type, it takes many years to test the current “estimation” of increased cancer risk in patients exposed to CT, a technology which has not been around long enough to permit such study in children. The reported risks are “estimates” calculated from a multitude of factors that are dependent upon radiation dose per organ as a function of gender, age, weight, type of CT, and dose calculation based on electronic models and data from studies of atomic bomb survivors. Nonetheless, it is certain that radiation from CT scans is a contributing factor to the rate of cancer in the general public, although to what extent is difficult to accurately determine. The ALARA conference organizers call for a more judicious selection of patients for CT examination and note the need for further investigation of appropriate parameters... You do not currently have access to this content.