Toward Precision Diagnosis: Machine Learning in Identifying Malignant Orbital Tumors With Multiparametric 3 T MRI.

BACKGROUND: Orbital tumors present a diagnostic challenge due to their varied locations and histopathological differences. Although recent advancements in imaging have improved diagnosis, classification remains a challenge. The integration of artificial intelligence in radiology and ophthalmology has demonstrated promising outcomes. PURPOSE: This study aimed to evaluate the performance of machine learning models in accurately distinguishing malignant orbital tumors from benign ones using multiparametric 3 T magnetic resonance imaging (MRI) data. MATERIALS AND METHODS: In this single-center prospective study, patients with orbital masses underwent presurgery 3 T MRI scans between December 2015 and May 2021. The MRI protocol comprised multiparametric imaging including dynamic contrast-enhanced (DCE), diffusion-weighted imaging (DWI), intravoxel incoherent motion (IVIM), as well as morphological imaging acquisitions. A repeated nested cross-validation strategy using random forest classifiers was used for model training and evaluation, considering 8 combinations of explanatory features. Shapley additive explanations (SHAP) values were used to assess feature contributions, and the model performance was evaluated using multiple metrics. RESULTS: One hundred thirteen patients were analyzed (57/113 [50.4%] were women; average age was 51.5 ± 17.5 years, range: 19-88 years). Among the 8 combinations of explanatory features assessed, the performance on predicting malignancy when using the most comprehensive model, which is the most exhaustive one incorporating all 46 explanatory features-including morphology, DWI, DCE, and IVIM, achieved an area under the curve of 0.9 [0.73-0.99]. When using the streamlined "10-feature signature" model, performance reached an area under the curve of 0.88 [0.71-0.99]. Random forest feature importance graphs measured by the mean of SHAP values pinpointed the 10 most impactful features, which comprised 3 quantitative IVIM features, 4 quantitative DCE features, 1 quantitative DWI feature, 1 qualitative DWI feature, and age. CONCLUSIONS: Our findings demonstrate that a machine learning approach, integrating multiparametric MRI data such as DCE, DWI, IVIM, and morphological imaging, offers high-performing models for differentiating malignant from benign orbital tumors. The streamlined 10-feature signature, with a performance close to the comprehensive model, may be more suitable for clinical application.

Diagnostic Performance of Dynamic Contrast-Enhanced 3T MR Imaging for Characterization of Orbital Lesions: Validation in a Large Prospective Study.

BACKGROUND AND PURPOSE: Orbital lesions are rare but serious. Their characterization remains challenging. Diagnosis is based on biopsy or surgery, which implies functional risks. It is necessary to develop noninvasive diagnostic tools. The goal of this study was to evaluate the diagnostic performance of dynamic contrast-enhanced MR imaging at 3T when distinguishing malignant from benign orbital tumors on a large prospective cohort. MATERIALS AND METHODS: This institutional review board-approved prospective single-center study enrolled participants presenting with an orbital lesion undergoing a 3T MR imaging before surgery from December 2015 to May 2021. Morphologic, diffusion-weighted, and dynamic contrast-enhanced MR images were assessed by 2 readers blinded to all data. Univariable and multivariable analyses were performed. To assess diagnostic performance, we used the following metrics: area under the curve, sensitivity, and specificity. Histologic analysis, obtained through biopsy or surgery, served as the criterion standard for determining the benign or malignant status of the tumor. RESULTS: One hundred thirty-one subjects (66/131 [50%] women and 65/131 [50%] men; mean age, 52 [SD, 17.1] years; range, 19-88 years) were enrolled. Ninety of 131 (69%) had a benign lesion, and 41/131 (31%) had a malignant lesion. Univariable analysis showed a higher median of transfer constant from blood plasma to the interstitial environment (K trans) and of transfer constant from the interstitial environment to the blood plasma (minute-1) (Kep) and a higher interquartile range of K trans in malignant-versus-benign lesions (1.1 minute-1 versus 0.65 minute-1, P = .03; 2.1 minute-1 versus 1.1 minute-1, P = .01; 0.81 minute-1 versus 0.65 minute-1, P = .009, respectively). The best-performing multivariable model in distinguishing malignant-versus-benign lesions included parameters from dynamic contrast-enhanced imaging, ADC, and morphology and reached an area under the curve of 0.81 (95% CI, 0.67-0.96), a sensitivity of 0.82 (95% CI, 0.55-1), and a specificity of 0.81 (95% CI, 0.65-0.96). CONCLUSIONS: Dynamic contrast-enhanced MR imaging at 3T appears valuable when characterizing orbital lesions and provides complementary information to morphologic imaging and DWI.

2015 UK software audit of hepatobiliary scintigraphy.

INTRODUCTION: This audit investigated hepatobiliary function imaging in UK hospitals, reviewing protocol differences in acquisition and processing parameters and the effect on calculated gallbladder ejection fraction (GBEF). PARTICIPANTS AND METHODS: Two dynamic data sets were available: one continuous dynamic data set, and the other with a 5-min break to administer the fatty stimulus. Participants used a set of 12 anonymized patient data sets most similar to their standard protocol calculating GBEF using their routine method. RESULTS: Fifty-two UK centres responded. Across all centres for all data sets, there was large variability in GBEF quoted, mostly owing to variations in the calculation method, motion correction and imaging type/times. The largest contributor to GBEF variation was time acquired after stimulus which varied from 20 to 70 min. Only 48.1% centres acquired for 60 min after stimulus, which is the acquisition time stated in normal range references. Overall, 13.5% participating centres administered fatty stimuli that fell below the recommended 10 g. Widespread variations were found in GBEF normal ranges and fatty stimulus administration. Motion correction has a large effect on GBEF; in one data set, motion correction alone changed GBEF from 44 to 9%, but 25% of the participants stated motion correction was not used. CONCLUSION: The authors proposed gold standard is fat content of the stimulus should be at least 10 g; and images should be acquired for 60 min after stimulus. If GBEF is quoted, motion correction should be used, and if compared with a normal range, the stimulus used must fit with the reference.

Discovery Molecular Imaging Digital Ready PET/CT performance evaluation according to the NEMA NU2-2012 standard.

OBJECTIVES: The aim of this study was to evaluate and benchmark the performance characteristics of the General Electric (GE) Discovery Molecular Imaging (MI) Digital Ready (DR) PET/CT. MATERIALS AND METHODS: Performance evaluation against the National Electrical Manufacturers Association (NEMA) 2012 standard was performed on three GE Discovery MI DR PET/CT systems installed across different UK centres. The Discovery MI DR performance was compared with the Siemens Biograph mCT Flow, Phillips Ingenuity TF and GE Discovery 690 fully analogue PET/CT systems. In addition, as the Discovery MI DR is upgradable to the Digital MI with silicon photomultipliers, performance characteristics between analogue and digital were compared with assess potential benefits of a system upgrade. RESULTS: The average NEMA results across three Discovery MI DR scanners were: sensitivity 7.3 cps/kBq, spatial resolution full-width-half-maximum radial 5.5 mm, tangential 4.5 mm and axial 6 mm at 10 cm from the centre of the field-of-view, peak noise equivalent count rate 142 kcps, scatter fraction 37.1%, contrast recovery coefficients from the International Electrotechnical Commission phantom ranged from 52 to 87% for 10-37-mm diameter spheres. CONCLUSION: All three Discovery MI DR systems tested in this study exceeded the manufacturer's NEMA specification, yet variability between scanners was noted. Discovery MI DR showed similar performance to Discovery 690 and Ingenuity TF, but lower sensitivity and spatial resolution than Biograph mCT Flow. The Discovery MI DR showed lower spatial resolution and contrast recovery than the 20-cm field-of-view Digital MI.

Reducing CT dose in myocardial perfusion SPECT/CT.

The aim of this study was to reduce the radiation dose arising from computed tomography (CT) attenuation correction to single photon emission computed tomography myocardial perfusion imaging studies without adversely affecting its accuracy. Using the Perspex CTDI phantom with the Xi detector to measure dose, CT scans were acquired using the Siemens Symbia T over the full range of CT settings available. Using the default setting 'AECmean', the measured dose at the centre of the phantom was 1.68 mGy and the breast dose from the scout view was 0.30 mGy. The lowest dose was achieved using the dose modulation setting in which the doses were reduced to 1.21 mGy and undetectable (<0.01 mGy), respectively. To observe the effect of changing these settings, 30 patients received a stress scan with default CT settings and a rest scan utilizing single photon emission computed tomography-guided CT and the dose modulation CT settings. Results showed a mean effective dose reduction of 23.6%. The dose reduction was greatest for larger patients, with the largest dose reduction for one patient being 72%. There was no apparent difference in attenuation correction between the two sets of resultant images. These new lower-dose settings are now applied to all clinical myocardial perfusion imaging studies.

UK audit of glomerular filtration rate measurement from plasma sampling in 2013.

INTRODUCTION: An audit was carried out into UK glomerular filtration rate (GFR) calculation. The results were compared with an identical 2001 audit. METHODS: Participants used their routine method to calculate GFR for 20 data sets (four plasma samples) in millilitres per minute and also the GFR normalized for body surface area. Some unsound data sets were included to analyse the applied quality control (QC) methods. Variability between centres was assessed for each data set, compared with the national median and a reference value calculated using the method recommended in the British Nuclear Medicine Society guidelines. The influence of the number of samples on variability was studied. Supplementary data were requested on workload and methodology. RESULTS: The 59 returns showed widespread standardization. The applied early exponential clearance correction was the main contributor to the observed variability. These corrections were applied by 97% of centres (50% - 2001) with 80% using the recommended averaged Brochner-Mortenson correction. Approximately 75% applied the recommended Haycock body surface area formula for adults (78% for children). The effect of the number of samples used was not significant. There was wide variability in the applied QC techniques, especially in terms of the use of the volume of distribution. CONCLUSION: The widespread adoption of the guidelines has harmonized national GFR calculation compared with the previous audit. Further standardization could further reduce variability. This audit has highlighted the need to address the national standardization of QC methods. Radionuclide techniques are confirmed as the preferred method for GFR measurement when an unequivocal result is required.

Accurate localization of a fall in pH within the ileocecal region: validation using a dual-scintigraphic technique.

Stereotypical changes in pH occur along the gastrointestinal (GI) tract. Classically, there is an abrupt increase in pH on exit from the stomach, followed later by a sharp fall in pH, attributed to passage through the ileocecal region. However, the precise location of this latter pH change has never been conclusively substantiated. We aimed to determine the site of fall in pH using a dual-scintigraphic technique. On day 1, 13 healthy subjects underwent nasal intubation with a 3-m-long catheter, which was allowed to progress to the distal ileum. On day 2, subjects ingested a pH-sensitive wireless motility capsule labeled with 4 MBq (51)Chromium [EDTA]. The course of this, as it travelled through the GI tract, was assessed with a single-headed γ-camera using static and dynamic scans. Capsule progression was plotted relative to a background of 4 MBq ¹¹¹Indium [diethylenetriamine penta-acetic acid] administered through the catheter. Intraluminal pH, as recorded by the capsule, was monitored continuously, and position of the capsule relative to pH was established. A sharp fall in pH was recorded in all subjects; position of the capsule relative to this was accurately determined anatomically in 9/13 subjects. In these nine subjects, a pH drop of 1.5 ± 0.2 U, from 7.6 ± 0.05 to 6.1 ± 0.1 occurred a median of 7.5 min (1-16) after passage through the ileocecal valve; location was either in the cecum (n = 5), ascending colon (n = 2), or coincident with a move from the cecum to ascending colon (n = 2). This study provides conclusive evidence that the fall in pH seen within the ileocolonic region actually occurs in the proximal colon. This phenomenon can be used as a biomarker of transition between the small and large bowel and validates assessment of regional GI motility using capsule technology that incorporates pH measurement.