Two is better than one: longitudinal detection and volumetric evaluation of brain metastases after Stereotactic Radiosurgery with a deep learning pipeline.

PURPOSE: Close MRI surveillance of patients with brain metastases following Stereotactic Radiosurgery (SRS) treatment is essential for assessing treatment response and the current disease status in the brain. This follow-up necessitates the comparison of target lesion sizes in pre- (prior) and post-SRS treatment (current) T1W-Gad MRI scans. Our aim was to evaluate SimU-Net, a novel deep-learning model for the detection and volumetric analysis of brain metastases and their temporal changes in paired prior and current scans. METHODS: SimU-Net is a simultaneous multi-channel 3D U-Net model trained on pairs of registered prior and current scans of a patient. We evaluated its performance on 271 pairs of T1W-Gad MRI scans from 226 patients who underwent SRS. An expert oncological neurosurgeon manually delineated 1,889 brain metastases in all the MRI scans (1,368 with diameters > 5 mm, 834 > 10 mm). The SimU-Net model was trained/validated on 205 pairs from 169 patients (1,360 metastases) and tested on 66 pairs from 57 patients (529 metastases). The results were then compared to the ground truth delineations. RESULTS: SimU-Net yielded a mean (std) detection precision and recall of 1.00±0.00 and 0.99±0.06 for metastases > 10 mm, 0.90±0.22 and 0.97±0.12 for metastases > 5 mm of, and 0.76±0.27 and 0.94±0.16 for metastases of all sizes. It improves lesion detection precision by 8% for all metastases sizes and by 12.5% for metastases < 10 mm with respect to standalone 3D U-Net. The segmentation Dice scores were 0.90±0.10, 0.89±0.10 and 0.89±0.10 for the above metastases sizes, all above the observer variability of 0.80±0.13. CONCLUSION: Automated detection and volumetric quantification of brain metastases following SRS have the potential to enhance the assessment of treatment response and alleviate the clinician workload.

Deep learning-based segmentation of whole-body fetal MRI and fetal weight estimation: assessing performance, repeatability, and reproducibility.

OBJECTIVES: To develop a deep-learning method for whole-body fetal segmentation based on MRI; to assess the method's repeatability, reproducibility, and accuracy; to create an MRI-based normal fetal weight growth chart; and to assess the sensitivity to detect fetuses with growth restriction (FGR). METHODS: Retrospective data of 348 fetuses with gestational age (GA) of 19-39 weeks were included: 249 normal appropriate for GA (AGA), 19 FGR, and 80 Other (having various imaging abnormalities). A fetal whole-body segmentation model with a quality estimation module was developed and evaluated in 169 cases. The method was evaluated for its repeatability (repeated scans within the same scanner, n = 22), reproducibility (different scanners, n = 6), and accuracy (compared with birth weight, n = 7). A normal MRI-based growth chart was derived. RESULTS: The method achieved a Dice = 0.973, absolute volume difference ratio (VDR) = 1.8% and VDR mean difference = 0.75% ([Formula: see text]: - 3.95%, 5.46), and high agreement with the gold standard. The method achieved a repeatability coefficient = 4.01%, ICC = 0.99, high reproducibility with a mean difference = 2.21% ([Formula: see text]: - 1.92%, 6.35%), and high accuracy with a mean difference between estimated fetal weight (EFW) and birth weight of - 0.39% ([Formula: see text]: - 8.23%, 7.45%). A normal growth chart (n = 246) was consistent with four ultrasound charts. EFW based on MRI correctly predicted birth-weight percentiles for all 18 fetuses ≤ 10thpercentile and for 14 out of 17 FGR fetuses below the 3rd percentile. Six fetuses referred to MRI as AGA were found to be < 3rd percentile. CONCLUSIONS: The proposed method for automatic MRI-based EFW demonstrated high performance and sensitivity to identify FGR fetuses. CLINICAL RELEVANCE STATEMENT: Results from this study support the use of the automatic fetal weight estimation method based on MRI for the assessment of fetal development and to detect fetuses at risk for growth restriction. KEY POINTS: • An AI-based segmentation method with a quality assessment module for fetal weight estimation based on MRI was developed, achieving high repeatability, reproducibility, and accuracy. • An MRI-based fetal weight growth chart constructed from a large cohort of normal and appropriate gestational-age fetuses is proposed. • The method showed a high sensitivity for the diagnosis of small fetuses suspected of growth restriction.

The anatomical pattern of ductal arborization in parotid glands using cone-beam computerized sialography.

Cone-Beam Computed Tomography-Sialography (Sialo-CBCT) is used to demonstrate salivary ductal structure. This study aimed to conduct a volumetric analysis of the anatomical morphology of Normal-Appearing Glands (NAGs) in parotid sialo-CBCT. Our retrospective study included 14 parotid sialo-CBCT scans interpreted as NAGs in 11 patients with salivary gland impairment. The main duct length and width, as well as number and width of secondary and tertiary ducts were manually evaluated. We found that the main parotid duct showed an average width of 1.39 mm, 1.15 mm, and 0.98 mm, for the proximal, middle and distal thirds, respectively. The arborization pattern showed approximately 20% more tertiary (average number 11.1 ± 2.7) than secondary ducts (average number 9.0 ± 2.4) and approximately 8% narrower tertiary ducts (average width 0.65 ± 0.11 mm) compared to the secondary ducts (average width 0.77 ± 0.14 mm). Our anatomical analysis of NAGs in parotid sialo-CBCT demonstrated progressive narrowing of the main duct and increasing arborization and decreasing lumen size starting from the primary to the tertiary ducts. This is the most updated study regarding the anatomy of the parotid glands as demonstrated in sialo-CBCT. Our results may provide clinicians with the basic information for understanding aberration from normal morphology, as seen in salivary gland pathologies as well facilitate planning of treatment strategies, such as minimally invasive sialo-endoscopies, commonly practiced today.

Real-time influence of intracellular acidification and Na+ /H+ exchanger inhibition on in-cell pyruvate metabolism in the perfused mouse heart: A 31 P-NMR and hyperpolarized 13 C-NMR study.

Disruption of acid-base balance is linked to various diseases and conditions. In the heart, intracellular acidification is associated with heart failure, maladaptive cardiac hypertrophy, and myocardial ischemia. Previously, we have reported that the ratio of the in-cell lactate dehydrogenase (LDH) to pyruvate dehydrogenase (PDH) activities is correlated with cardiac pH. To further characterize the basis for this correlation, these in-cell activities were investigated under induced intracellular acidification without and with Na+ /H+ exchanger (NHE1) inhibition by zoniporide. Male mouse hearts (n = 30) were isolated and perfused retrogradely. Intracellular acidification was performed in two ways: (1) with the NH4 Cl prepulse methodology; and (2) by combining the NH4 Cl prepulse with zoniporide. 31 P NMR spectroscopy was used to determine the intracellular cardiac pH and to quantify the adenosine triphosphate and phosphocreatine content. Hyperpolarized [1-13 C]pyruvate was obtained using dissolution dynamic nuclear polarization. 13 C NMR spectroscopy was used to monitor hyperpolarized [1-13 C]pyruvate metabolism and determine enzyme activities in real time at a temporal resolution of a few seconds using the product-selective saturating excitation approach. The intracellular acidification induced by the NH4 Cl prepulse led to reduced LDH and PDH activities (-16% and -39%, respectively). This finding is in line with previous evidence of reduced myocardial contraction and therefore reduced metabolic activity upon intracellular acidification. Concomitantly, the LDH/PDH activity ratio increased with the reduction in pH, as previously reported. Combining the NH4 Cl prepulse with zoniporide led to a greater reduction in LDH activity (-29%) and to increased PDH activity (+40%). These changes resulted in a surprising decrease in the LDH/PDH ratio, as opposed to previous predictions. Zoniporide alone (without intracellular acidification) did not change these enzyme activities. A possible explanation for the enzymatic changes observed during the combination of the NH4 Cl prepulse and NHE1 inhibition may be related to mitochondrial NHE1 inhibition, which likely negates the mitochondrial matrix acidification. This effect, combined with the increased acidity in the cytosol, would result in an enhanced H+ gradient across the mitochondrial membrane and a temporarily higher pyruvate transport into the mitochondria, thereby increasing the PDH activity at the expense of the cytosolic LDH activity. These findings demonstrate the complexity of in-cell cardiac metabolism and its dependence on intracellular acidification. This study demonstrates the capabilities and limitations of hyperpolarized [1-13 C]pyruvate in the characterization of intracellular acidification as regards cardiac pathologies.

Fetal MRI-Based Body and Adiposity Quantification for Small for Gestational Age Perinatal Risk Stratification.

BACKGROUND: Small for gestational age (SGA) fetuses are at risk for perinatal adverse outcomes. Fetal body composition reflects the fetal nutrition status and hold promise as potential prognostic indicator. MRI quantification of fetal anthropometrics may enhance SGA risk stratification. HYPOTHESIS: Smaller, leaner fetuses are malnourished and will experience unfavorable outcomes. STUDY TYPE: Prospective. POPULATION: 40 SGA fetuses, 26 (61.9%) females: 10/40 (25%) had obstetric interventions due to non-reassuring fetal status (NRFS), and 17/40 (42.5%) experienced adverse neonatal events (CANO). Participants underwent MRI between gestational ages 30 + 2 and 37 + 2. FIELD STRENGTH/SEQUENCE: 3-T, True Fast Imaging with Steady State Free Precession (TruFISP) and T1 -weighted two-point Dixon (T1 W Dixon) sequences. ASSESSMENT: Total body volume (TBV), fat signal fraction (FSF), and the fat-to-body volumes ratio (FBVR) were extracted from TruFISP and T1 W Dixon images, and computed from automatic fetal body and subcutaneous fat segmentations by deep learning. Subjects were followed until hospital discharge, and obstetric interventions and neonatal adverse events were recorded. STATISTICAL TESTS: Univariate and multivariate logistic regressions for the association between TBV, FBVR, and FSF and interventions for NRFS and CANO. Fisher's exact test was used to measure the association between sonographic FGR criteria and perinatal outcomes. Sensitivity, specificity, positive and negative predictive values, and accuracy were calculated. A P-value <0.05 was considered statistically significant. RESULTS: FBVR (odds ratio [OR] 0.39, 95% confidence interval [CI] 0.2-0.76) and FSF (OR 0.95, CI 0.91-0.99) were linked with NRFS interventions. Furthermore, TBV (OR 0.69, CI 0.56-0.86) and FSF (OR 0.96, CI 0.93-0.99) were linked to CANO. The FBVR sensitivity/specificity for obstetric interventions was 85.7%/87.5%, and the TBV sensitivity/specificity for CANO was 82.35%/86.4%. The sonographic criteria sensitivity/specificity for obstetric interventions was 100%/33.3% and insignificant for CANO (P = 0.145). DATA CONCLUSION: Reduced TBV and FBVR may be associated with higher rates of obstetric interventions for NRFS and CANO. EVIDENCE LEVEL: 2 TECHNICAL EFFICACY: Stage 5.

Follow-up of liver metastases: a comparison of deep learning and RECIST 1.1.

OBJECTIVES: To compare liver metastases changes in CT assessed by radiologists using RECIST 1.1 and with aided simultaneous deep learning-based volumetric lesion changes analysis. METHODS: A total of 86 abdominal CT studies from 43 patients (prior and current scans) of abdominal CT scans of patients with 1041 liver metastases (mean = 12.1, std = 11.9, range 1-49) were analyzed. Two radiologists performed readings of all pairs; conventional with RECIST 1.1 and with computer-aided assessment. For computer-aided reading, we used a novel simultaneous multi-channel 3D R2U-Net classifier trained and validated on other scans. The reference was established by having an expert radiologist validate the computed lesion detection and segmentation. The results were then verified and modified as needed by another independent radiologist. The primary outcome measure was the disease status assessment with the conventional and the computer-aided readings with respect to the reference. RESULTS: For conventional and computer-aided reading, there was a difference in disease status classification in 40 out of 86 (46.51%) and 10 out of 86 (27.9%) CT studies with respect to the reference, respectively. Computer-aided reading improved conventional reading in 30 CT studies by 34.5% for two readers (23.2% and 46.51%) with respect to the reference standard. The main reason for the difference between the two readings was lesion volume differences (p = 0.01). CONCLUSIONS: AI-based computer-aided analysis of liver metastases may improve the accuracy of the evaluation of neoplastic liver disease status. CLINICAL RELEVANCE STATEMENT: AI may aid radiologists to improve the accuracy of evaluating changes over time in metastasis of the liver. KEY POINTS: • Classification of liver metastasis changes improved significantly in one-third of the cases with an automatically generated comprehensive lesion and lesion changes report. • Simultaneous deep learning changes detection and volumetric assessment may improve the evaluation of liver metastases temporal changes potentially improving disease management.

Improved differentiation between hypo/hypertelorism and normal fetuses based on MRI using automatic ocular biometric measurements, ocular ratios, and machine learning multi-parametric classification.

OBJECTIVES: To differentiate hypo-/hypertelorism (abnormal) from normal fetuses using automatic biometric measurements and machine learning (ML) classification based on MRI. METHODS: MRI data of normal (n = 244) and abnormal (n = 52) fetuses of 22-40 weeks' gestational age (GA), scanned between March 2008 and June 2020 on 1.5/3T systems with various T2-weighted sequences and image resolutions, were included. A fully automatic method including deep learning and geometric algorithms was developed to measure the binocular (BOD), inter-ocular (IOD), ocular (OD) diameters, and ocular volume (OV). Two new parameters, BOD-ratio and IOD-ratio, were defined as the ratio between BOD/IOD relative to the sum of both globes' OD, respectively. Eight ML classifiers were evaluated to detect abnormalities using measured and computed parameters. RESULTS: The automatic method yielded a mean difference of BOD = 0.70 mm, IOD = 0.81 mm, OD = 1.00 mm, and a 3D-Dice score of OV = 93.7%. In normal fetuses, all four measurements increased with GA. Constant values were detected for BOD-ratio = 1.56 ± 0.05 and IOD-ratio = 0.60 ± 0.05 across all GA and when calculated from previously published reference data of both MRI and ultrasound. A random forest classifier yielded the best results on an independent test set (n = 58): AUC-ROC = 0.941 and F1-Score = 0.711 in comparison to AUC-ROC = 0.650 and F1-Score = 0.385 achieved based on the accepted criteria that define hypo/hypertelorism based on IOD (< 5th or > 95th percentiles). Using the explainable ML method, the two computed ratios were found as the most contributing parameters. CONCLUSIONS: The developed fully automatic method demonstrates high performance on varied clinical imaging data. The new BOD and IOD ratios and ML multi-parametric classifier are suggested to improve the differentiation of hypo-/hypertelorism from normal fetuses. KEY POINTS: • A fully automatic method for computing fetal ocular biometry from MRI is proposed, achieving high performance, comparable to that of an expert fetal neuro-radiologist. • Two new parameters, IOD-ratio and BOD-ratio, are proposed for routine clinical use in ultrasound and MRI. These two ratios are constant across gestational age in normal fetuses, consistent across studies, and differentiate between fetuses with and without hypo/hypertelorism. • Multi-parametric machine learning classification based on automatic measurements and the two new ratios improves the identification of fetal ocular anomalies beyond the accepted criteria (<5th or >95th IOD percentiles).

Reduced adipose tissue in growth-restricted fetuses using quantitative analysis of magnetic resonance images.

OBJECTIVES: Fat-water MRI can be used to quantify tissues' lipid content. We aimed to quantify fetal third trimester normal whole-body subcutaneous lipid deposition and explore differences between appropriate for gestational age (AGA), fetal growth restriction (FGR), and small for gestational age fetuses (SGAs). METHODS: We prospectively recruited women with FGR and SGA-complicated pregnancies and retrospectively recruited the AGA cohort (sonographic estimated fetal weight [EFW] ≥ 10th centile). FGR was defined using the accepted Delphi criteria, and fetuses with an EFW < 10th centile that did not meet the Delphi criteria were defined as SGA. Fat-water and anatomical images were acquired in 3 T MRI scanners. The entire fetal subcutaneous fat was semi-automatically segmented. Three adiposity parameters were calculated: fat signal fraction (FSF) and two novel parameters, i.e., fat-to-body volume ratio (FBVR) and estimated total lipid content (ETLC = FSF*FBVR). Normal lipid deposition with gestation and differences between groups were assessed. RESULTS: Thirty-seven AGA, 18 FGR, and 9 SGA pregnancies were included. All three adiposity parameters increased between 30 and 39 weeks (p < 0.001). All three adiposity parameters were significantly lower in FGR compared with AGA (p ≤ 0.001). Only ETLC and FSF were significantly lower in SGA compared with AGA using regression analysis (p = 0.018-0.036, respectively). Compared with SGA, FGR had a significantly lower FBVR (p = 0.011) with no significant differences in FSF and ETLC (p ≥ 0.053). CONCLUSIONS: Whole-body subcutaneous lipid accretion increased throughout the third trimester. Reduced lipid deposition is predominant in FGR and may be used to differentiate FGR from SGA, assess FGR severity, and study other malnourishment pathologies. CLINICAL RELEVANCE STATEMENT: Fetuses with growth restriction have reduced lipid deposition than appropriately developing fetuses measured using MRI. Reduced fat accretion is linked with worse outcomes and may be used for growth restriction risk stratification. KEY POINTS: • Fat-water MRI can be used to assess the fetal nutritional status quantitatively. • Lipid deposition increased throughout the third trimester in AGA fetuses. • FGR and SGA have reduced lipid deposition compared with AGA fetuses, more predominant in FGR.

Contribution of information about acute and geriatric characteristics to decisions about life-sustaining treatment for old patients in intensive care.

BACKGROUND: Life-sustaining treatment (LST) in the intensive care unit (ICU) is withheld or withdrawn when there is no reasonable expectation of beneficial outcome. This is especially relevant in old patients where further functional decline might be detrimental for the self-perceived quality of life. However, there still is substantial uncertainty involved in decisions about LST. We used the framework of information theory to assess that uncertainty by measuring information processed during decision-making. METHODS: Datasets from two multicentre studies (VIP1, VIP2) with a total of 7488 ICU patients aged 80 years or older were analysed concerning the contribution of information about the acute illness, age, gender, frailty and other geriatric characteristics to decisions about LST. The role of these characteristics in the decision-making process was quantified by the entropy of likelihood distributions and the Kullback-Leibler divergence with regard to withholding or withdrawing decisions. RESULTS: Decisions to withhold or withdraw LST were made in 2186 and 1110 patients, respectively. Both in VIP1 and VIP2, information about the acute illness had the lowest entropy and largest Kullback-Leibler divergence with respect to decisions about withdrawing LST. Age, gender and geriatric characteristics contributed to that decision only to a smaller degree. CONCLUSIONS: Information about the severity of the acute illness and, thereby, short-term prognosis dominated decisions about LST in old ICU patients. The smaller contribution of geriatric features suggests persistent uncertainty about the importance of functional outcome. There still remains a gap to fully explain decision-making about LST and further research involving contextual information is required. TRIAL REGISTRATION: VIP1 study: NCT03134807 (1 May 2017), VIP2 study: NCT03370692 (12 December 2017).

Computer-Generated Radiographic Measurements of Distal Radius Fractures: Does It Help With Decision Making?

PURPOSE: Operative management of distal radius fractures (DRFs) has become increasingly common. Age, activity levels, and comorbid conditions are major factors influencing the treatment decision, although operative indications are still controversial. Radiographic parameters (RPs), such as radial inclination, dorsal tilt, and articular step-off, can provide objective support for effective decision making. However, manual measurement of RPs may be imprecise and subject to inconsistency. To address this problem, we developed custom software of an algorithm to automatically detect and compute 6 common RPs associated with DRF in anteroposterior and lateral radiographs. The aim in this study was to assess the effect of this software on radiographic interobserver variability among orthopedic surgeons. Our hypothesis was that precise and consistent measurement of RPs will improve radiographic interpretation variability among surgeons and, consequently, may aid in clinical decision making. METHODS: Thirty-five radiograph series of DRFs were presented to 9 fellowship-trained hand and orthopedic trauma surgeons. Each case was presented with basic clinical information, together with plain anteroposterior and lateral radiographs. One of the 2 possible treatment options was selected: casting or open reduction with a locking plate. The survey was repeated 3 weeks later, this time with computer-generated RP measurements. Data were analyzed for interobserver and intraobserver variability for both surveys, and the interclass coefficient, kappa value, was calculated. RESULTS: The interobserver reliability (interclass coefficient value) improved from poor to moderate, 0.35 to 0.50, with the provided RP. The average intraobserver interclass coefficient was 0.68. When participants were assessed separately according to their subspecialties (trauma and hand), improved interobserver variability was found as well. CONCLUSIONS: Providing computed RPs to orthopedic surgeons may improve the consistency of the radiographic judgment and influence their clinical decision for the treatment of DRFs. CLINICAL RELEVANCE: Orthopedic surgeons' consistency in the radiographic judgment of DRFs slightly improved by providing automatically calculated radiographic measurements to them.

Change detection in sparse repeat CT scans with non-rigid deformations.

BACKGROUND: Detecting and interpreting changes in the images of follow-up CT scans by the clinicians is often time-consuming and error-prone due to changes in patient position and non-rigid anatomy deformations. Thus, reconstructed repeat scan images are required, precluding reduced dose sparse-view repeat scanning. OBJECTIVE: A method to automatically detect changes in a region of interest of sparse-view repeat CT scans in the presence of non-rigid deformations of the patient's anatomy without reconstructing the original images. METHODS: The proposed method uses the sparse sinogram data of two CT scans to distinguish between genuine changes in the repeat scan and differences due to non-rigid anatomic deformations. First, size and contrast level of the changed regions are estimated from the difference between the scans' sinogram data. The estimated types of changes in the repeat scan help optimize the method's parameter values. Two scans are then aligned using Radon space non-rigid registration. Rays which crossed changes in the ROI are detected and back-projected onto image space in a two-phase procedure. These rays form a likelihood map from which the binary changed region map is computed. RESULTS: Experimental studies on four pairs of clinical lung and liver CT scans with simulated changed regions yield a mean changed region recall rate > 86%and a mean precision rate > 83%when detecting large changes with low contrast, and high contrast changes, even when small. The new method outperforms image space methods using prior image constrained compressed sensing (PICCS) reconstruction, particularly for small, low contrast changes (recall = 15.8%, precision = 94.7%). CONCLUSION: Our method for automatic change detection in sparse-view repeat CT scans with non-rigid deformations may assist radiologists by highlighting the changed regions and may obviate the need for a high-quality repeat scan image when no changes are detected.

Deformable registration and region-of-interest image reconstruction in sparse repeat CT scanning.

BACKGROUND: Repeat CT scanning is ubiquitous in many clinical situations, e.g. to follow disease progression, to evaluate treatment efficacy, and to monitor interventional CT procedures. However, it incurs in cumulative radiation to the patient which can be significantly reduced by using a region of interest (ROI) and the existing baseline scan. OBJECTIVE: To obtain a high-quality reconstruction of a ROI with a significantly reduced X-ray radiation dosage that accounts for deformations. METHODS: We present a new method for deformable registration and image reconstruction inside an ROI in repeat CT scans with a highly reduced X-ray radiation dose based on sparse scanning. Our method uses the existing baseline scan data, a user-defined ROI, and a new sparse repeat scan to compute a high-quality repeat scan ROI image with a significantly reduced radiation dose. Our method first performs rigid registration between the densely scanned baseline and the sparsely scanned repeat CT scans followed by deformable registration with a low-order parametric model, both in 3D Radon space and without reconstructing the repeat scan image. It then reconstructs the repeat scan ROI without computing the entire repeat scan image. RESULTS: Our experimental results on clinical lung and liver CT scans yield a mean × 14 computation speedup and a × 7.6-12.5 radiation dose reduction, with a minor image quality loss of 0.0157 in the NRMSE metric. CONCLUSION: Our method is considerably faster than existing methods, thereby enabling intraoperative online repeat scanning that it is accurate and accounts for position, deformation, and structure changes at a fraction of the radiation dose required by existing methods.

Inter-observer variability of manual contour delineation of structures in CT.

PURPOSE: To quantify the inter-observer variability of manual delineation of lesions and organ contours in CT to establish a reference standard for volumetric measurements for clinical decision making and for the evaluation of automatic segmentation algorithms. MATERIALS AND METHODS: Eleven radiologists manually delineated 3193 contours of liver tumours (896), lung tumours (1085), kidney contours (434) and brain hematomas (497) on 490 slices of clinical CT scans. A comparative analysis of the delineations was then performed to quantify the inter-observer delineation variability with standard volume metrics and with new group-wise metrics for delineations produced by groups of observers. RESULTS: The mean volume overlap variability values and ranges (in %) between the delineations of two observers were: liver tumours 17.8 [-5.8,+7.2]%, lung tumours 20.8 [-8.8,+10.2]%, kidney contours 8.8 [-0.8,+1.2]% and brain hematomas 18 [-6.0,+6.0] %. For any two randomly selected observers, the mean delineation volume overlap variability was 5-57%. The mean variability captured by groups of two, three and five observers was 37%, 53% and 72%; eight observers accounted for 75-94% of the total variability. For all cases, 38.5% of the delineation non-agreement was due to parts of the delineation of a single observer disagreeing with the others. No statistical difference was found for the delineation variability between the observers based on their expertise. CONCLUSION: The variability in manual delineations for different structures and observers is large and spans a wide range across a variety of structures and pathologies. Two and even three observers may not be sufficient to establish the full range of inter-observer variability. KEY POINTS: • This study quantifies the inter-observer variability of manual delineation of lesions and organ contours in CT. • The variability of manual delineations between two observers can be significant. Two and even three observers capture only a fraction of the full range of inter-observer variability observed in common practice. • Inter-observer manual delineation variability is necessary to establish a reference standard for radiologist training and evaluation and for the evaluation of automatic segmentation algorithms.

Computer-Aided Orthopedic Surgery: Incremental Shift or Paradigm Change?

Computer-aided orthopedic surgery (CAOS) is now about 25 years old. Unlike neurosurgery, computer-aided surgery has not become the standard of care in orthopedic surgery. In this paper, we provide the technical and clinical context raised by this observation in an attempt to elucidate the reasons for this state of affairs. We start with a brief outline of the history of CAOS, review the main CAOS technologies, and describe how they are evaluated. We then identify some of the current publications in the field and present the opposing views on their clinical impact and their acceptance by the orthopedic community worldwide. We focus on total knee replacement surgery as a case study and present current clinical results and contrasting opinions on CAOS technologies. We then discuss the challenges and opportunities for research in medical image analysis in CAOS and in musculoskeletal radiology. We conclude with a suggestion that while CAOS acceptance may be more moderate than that of other fields in surgery, it still has a place in the arsenal of useful tools available to orthopedic surgeons.

Automatic Measurement of Fetal Brain Development from Magnetic Resonance Imaging: New Reference Data.

BACKGROUND: Accurate fetal brain volume estimation is of paramount importance in evaluating fetal development. The aim of this study was to develop an automatic method for fetal brain segmentation from magnetic resonance imaging (MRI) data, and to create for the first time a normal volumetric growth chart based on a large cohort. SUBJECTS AND METHODS: A semi-automatic segmentation method based on Seeded Region Growing algorithm was developed and applied to MRI data of 199 typically developed fetuses between 18 and 37 weeks' gestation. The accuracy of the algorithm was tested against a sub-cohort of ground truth manual segmentations. A quadratic regression analysis was used to create normal growth charts. The sensitivity of the method to identify developmental disorders was demonstrated on 9 fetuses with intrauterine growth restriction (IUGR). RESULTS: The developed method showed high correlation with manual segmentation (r2 = 0.9183, p < 0.001) as well as mean volume and volume overlap differences of 4.77 and 18.13%, respectively. New reference data on 199 normal fetuses were created, and all 9 IUGR fetuses were at or below the third percentile of the normal growth chart. DISCUSSION: The proposed method is fast, accurate, reproducible, user independent, applicable with retrospective data, and is suggested for use in routine clinical practice.

The influence of varying the number of characters per row on the accuracy and reproducibility of the ETDRS visual acuity chart.

BACKGROUND: As part of an effort to improve upon the Snellen chart, we provide a standardized version of the ETDRS chart utilizing five characters in each row. The choice of five characters contradicts the recommended ten characters per row determined by the NAS-NRC, a committee established to provide guidelines for testing visual acuity. We set out to quantify the influence of varying the number of characters per line on the ETDRS chart with respect to the accuracy and reproducibility of visual acuity measurement. METHODS: Eleven different ETDRS charts were created, each with a different number of characters appearing in each row. A computer simulation was programmed to run 10,000 virtual patients, each with a unique visual acuity, false-positive and false-negative error value. RESULTS: Accuracy and reproducibility were found to roughly correlate with the number of characters present in each row, such that charts with 1, 3, 5, 7, 9, and 11 characters per row provided accuracy of 0.164, 0.094, 0.078, 0.073, 0.071, and 0.070 logMAR, respectively. A non-linear relationship was observed, with little improvement found beyond seven characters per row. In addition, charts with an even number of characters per row provided higher accuracy than their greater-number odd counterparts. In certain instances, accuracy and reproducibility were not well correlated. CONCLUSIONS: Increasing the number of characters per row in the ETDRS chart provides a trade-off between accuracy and test duration. An optimized chart layout would take these findings into account, allowing for the use of different chart layouts for clinical versus research settings.

Current state of computer navigation and robotics in unicompartmental and total knee arthroplasty: a systematic review with meta-analysis.

Recently, there is a growing interest in surgical variables that are intraoperatively controlled by orthopaedic surgeons, including lower leg alignment, component positioning and soft tissues balancing. Since more tight control over these factors is associated with improved outcomes of unicompartmental knee arthroplasty and total knee arthroplasty (TKA), several computer navigation and robotic-assisted systems have been developed. Although mechanical axis accuracy and component positioning have been shown to improve with computer navigation, no superiority in functional outcomes has yet been shown. This could be explained by the fact that many differences exist between the number and type of surgical variables these systems control. Most systems control lower leg alignment and component positioning, while some in addition control soft tissue balancing. Finally, robotic-assisted systems have the additional advantage of improving surgical precision. A systematic search in PubMed, Embase and Cochrane Library resulted in 40 comparative studies and three registries on computer navigation reporting outcomes of 474,197 patients, and 21 basic science and clinical studies on robotic-assisted knee arthroplasty. Twenty-eight of these comparative computer navigation studies reported Knee Society Total scores in 3504 patients. Stratifying by type of surgical variables, no significant differences were noted in outcomes between surgery with computer-navigated TKA controlling for alignment and component positioning versus conventional TKA (p = 0.63). However, significantly better outcomes were noted following computer-navigated TKA that also controlled for soft tissue balancing versus conventional TKA (mean difference 4.84, 95 % Confidence Interval 1.61, 8.07, p = 0.003). A literature review of robotic systems showed that these systems can, similarly to computer navigation, reliably improve lower leg alignment, component positioning and soft tissues balancing. Furthermore, two studies comparing robotic-assisted with computer-navigated surgery reported superiority of robotic-assisted surgery in controlling these factors. Manually controlling all these surgical variables can be difficult for the orthopaedic surgeon. Findings in this study suggest that computer navigation or robotic assistance may help managing these multiple variables and could improve outcomes. Future studies assessing the role of soft tissue balancing in knee arthroplasty and long-term follow-up studies assessing the role of computer-navigated and robotic-assisted knee arthroplasty are needed.

The effect of chemotherapy on optic pathway gliomas and their sub-components: A volumetric MR analysis study.

BACKGROUND: Optic pathway gliomas (OPG) represent 5% of pediatric brain tumors and compose a major therapeutic dilemma to the treating physicians. While chemotherapy is widely used for these tumors, our ability to predict radiological response is still lacking. In this study, we use volumetric imaging to examine in detail the long-term effect of chemotherapy on the tumor as well as its various sub-components. PROCEDURE: The tumors of 15 patients with OPG, treated with chemotherapy, were longitudinally measured using our novel, previously described volumetric method. Patients were treated with up to five lines of chemotherapy. Sufficient follow-up imaging data, and patient's numbers, allowed for analysis of two treatment lines. Volumetric measurements of the tumors were segmented into solid-non-enhancing, solid-enhancing, and cystic components. Outcome analysis was done per specific treatment line and for the overall follow-up period. RESULTS: An average reduction of 9.7% (±23%) in the gross-total-solid volume (GTSV) was noted following treatment with vincristine and carboplatin. The cystic component grew under therapy by an average of 12.6% (±39%). When measured over the course of the whole study period, the cystic component grew by an average of 35% (±100%) and the GTSV increased by 12% (±35%). CONCLUSION: Initial treatment with vincristine and carboplatin seems to have a minimal initial effect, mostly on the solid components. The cystic component in itself seems to be unaffected by chemotherapy, and contributes to the subsequent growth of the total volume. During the overall treatment period, both solid and cystic components grew regardless of combined treatment methods.

Quantitative functional MRI biomarkers improved early detection of colorectal liver metastases.

PURPOSE: To implement and evaluate the performance of a computerized statistical tool designed for robust and quantitative analysis of hemodynamic response imaging (HRI) -derived maps for the early identification of colorectal liver metastases (CRLM). MATERIALS AND METHODS: CRLM-bearing mice were scanned during the early stage of tumor growth and subsequently during the advanced-stage. Three experienced radiologists marked various suspected-foci on the early stage anatomical images and classified each as either highly certain or as suspected tumors. The statistical model construction was based on HRI maps (functional-MRI combined with hypercapnia and hyperoxia) using a supervised learning paradigm which was further trained either with the advanced-stage sets (late training; LT) or with the early stage sets (early training; ET). For each group of foci, the classifier results were compared with the ground-truth. RESULTS: The ET-based classification significantly improved the manual classification of the highly certain foci (P < 0.05) and was superior compared with the LT-based classification (P < 0.05). Additionally, the ET-based classification, offered high sensitivity (57-63%), accompanied with high positive predictive value (>94%) and high specificity (>98%) for suspected-foci. CONCLUSION: The ET-based classifier can strengthen the radiologist's classification of highly certain foci. Additionally, it can aid in classifying suspected-foci, thus enabling earlier intervention which can often be lifesaving.

Fragment distance-guided dual-stream learning for automatic pelvic fracture segmentation.

Pelvic fracture is a complex and severe injury. Accurate diagnosis and treatment planning require the segmentation of the pelvic structure and the fractured fragments from preoperative CT scans. However, this segmentation is a challenging task, as the fragments from a pelvic fracture typically exhibit considerable variability and irregularity in the morphologies, locations, and quantities. In this study, we propose a novel dual-stream learning framework for the automatic segmentation and category labeling of pelvic fractures. Our method uniquely identifies pelvic fracture fragments in various quantities and locations using a dual-branch architecture that leverages distance learning from bone fragments. Moreover, we develop a multi-size feature fusion module that adaptively aggregates features from diverse receptive fields tailored to targets of different sizes and shapes, thus boosting segmentation performance. Extensive experiments on three pelvic fracture datasets from different medical centers demonstrated the accuracy and generalizability of the proposed method. It achieves a mean Dice coefficient and mean Sensitivity of 0.935±0.068 and 0.929±0.058 in the dataset FracCLINIC, and 0.955±0.072 and 0.912±0.125 in the dataset FracSegData, which are superior than other comparing methods. Our method optimizes the process of pelvic fracture segmentation, potentially serving as an effective tool for preoperative planning in the clinical management of pelvic fractures.