Radiology - Research Publications

Permanent URI for this collection

Search Results

Now showing 1 - 10 of 347
  • Item
    Thumbnail Image
    Verbal notification of radiology results: are radiologists meeting expectations?
    Preece, E ; Whitchurch, M ; Sutherland, T (Wiley, 2022-08)
    BACKGROUND: Delayed communication of radiographic findings is associated with poor patient outcomes and significant medicolegal risk. Radiologists verbally contact referring practitioners with urgent findings, although practitioner's expectations regarding notification have rarely been examined. AIM: To assess differences in preferred practice between radiologists and referring practitioners in the verbal communication of urgent radiology findings. METHODS: For 33 clinical stems, respondents were asked if they would issue (radiologists) or expect to receive (referring practitioners) verbal notification of results or routine written communication only. Surveys were emailed to radiologists and referring practitioners of varying experience at a tertiary referral hospital in Melbourne, Victoria. RESULTS: A total of 97 survey responses was received. Eighty responses were from referring practitioners and 17 from radiologists. Referring practitioners were seen to slightly prefer verbal notification more often than issued by radiologists overall (61%; 95% confidence interval (CI) 57-66% verbal notification expected vs 58%; 95% CI 52-64% issued). More senior referring practitioners with greater than 10 years' experience expected verbal notification more often (67%; 95% CI 59-75%), and more senior radiologists issued verbal reports less often (54%; 95% CI 39-69%). More junior referring practitioners, for example, registrars or fellows, expected notification less often overall (59%; 95% CI 43-76%). Subgroup analysis demonstrated statistically significant differences in notification preferences for certain clinical scenarios. CONCLUSIONS: Overall results show fair correlation between referrer's expectations of verbal notification and the provision of verbal notification by radiologists. However, there were discrepancies in the practice and preferences of more junior and senior practitioners in certain clinical scenarios.
  • Item
    Thumbnail Image
    Macroscopic fat containing renal cell carcinoma
    Kirkinis, M ; Sutherland, T (WILEY, 2021-03-05)
    Renal masses containing macroscopic fat traditionally are pathognomonic for angiomyolipoma, a benign tumour. We describe two cases contrary to this axiom, the first being initially referred for angioembolisation, but subsequently biopsied when it was angiographically occult, whilst the second case showed a small macroscopic fat component and arterial enhancement prompting biopsy. Neither of these two cases demonstrated calcification which would usually suggest a more sinister lesion requiring further workup. The results demonstrated renal cell carcinoma for both lesions. Our multidisciplinary meeting approach to renal masses with a small amount of macroscopic fat and no calcifications has now changed.
  • Item
    No Preview Available
    Computer-aided Measurement System Using Image Processing for Measuring Cobb Angle in Scoliosis
    Moftian, N ; Soltani, TS ; Salahzadeh, Z ; Pourfeizi, HH ; Gheibi, Y ; Fazlollahi, A ; Rezaei-Hachesu, P (Briefland, 2022-01-01)
    Background: One of the spine deformities is scoliosis, and Cobb angle is generally used to assess it. Objectives: In this study, a computer-aided measurement system (CAMS) was presented as a new repeatable and reproducible approach to assess the Cobb angle in idiopathic scoliosis patients. Methods: Python libraries, including OpenCV and Numpy were used for image processing and design of the software. To assess the repeatability and reproducibility of the CAMS, a series of 98 anterior-posterior radiographs from patients with idiopathic scoliosis were used. Assessments were done by five independent observers. Each radiograph was assessed by each observer three times with a minimum break of two weeks among assessment. The single measure intraclass correlation coefficient (ICC), the mean absolute difference (MAD), and the standard error measurement (SEM) values were used for intra- and inter-observer reliability. Results: The inter-observer analysis indicated that the ICCs ranged from 0.94 - 0.99, and the MAD between manual and CAMS were less than 3°. For intra-observer measurements, the combined SEM between all observers for the manual and CAMS was 1.79° and 1.27°, respectively. An ICC value of 0.97 with 95% confidence interval (CI) was excellent in CAMS for inter-observer reliability. The MAD of CAMS was 2.18 ± 2.01 degrees. Conclusions: The CAMS is an effective and reliable approach for assessing scoliotic curvature in the standing radiographs of thoraco-lumbar. Moreover, CAMS can accelerate clinical visits, and its calculation results are reliable.
  • Item
    No Preview Available
    Topographic divergence of atypical cortical asymmetry and atrophy patterns in temporal lobe epilepsy
    Park, B-Y ; Lariviere, S ; Rodriguez-Cruces, R ; Royer, J ; Tavakol, S ; Wang, Y ; Caciagli, L ; Caligiuri, ME ; Gambardella, A ; Concha, L ; Keller, SS ; Cendes, F ; Alvim, MKM ; Yasuda, C ; Bonilha, L ; Gleichgerrcht, E ; Focke, NK ; Kreilkamp, BAK ; Domin, M ; von Podewils, F ; Langner, S ; Rummel, C ; Rebsamen, M ; Wiest, R ; Martin, P ; Kotikalapudi, R ; Bender, B ; O'Brien, TJ ; Law, M ; Sinclair, B ; Vivash, L ; Kwan, P ; Desmond, PM ; Malpas, CB ; Lui, E ; Alhusaini, S ; Doherty, CP ; Cavalleri, GL ; Delanty, N ; Kalviainen, R ; Jackson, GD ; Kowalczyk, M ; Mascalchi, M ; Semmelroch, M ; Thomas, RH ; Soltanian-Zadeh, H ; Davoodi-Bojd, E ; Zhang, J ; Lenge, M ; Guerrini, R ; Bartolini, E ; Hamandi, K ; Foley, S ; Weber, B ; Depondt, C ; Absil, J ; Carr, SJA ; Abela, E ; Richardson, MP ; Devinsky, O ; Severino, M ; Striano, P ; Parodi, C ; Tortora, D ; Hatton, SN ; Vos, SB ; Duncan, JS ; Galovic, M ; Whelan, CD ; Bargallo, N ; Pariente, J ; Conde-Blanco, E ; Vaudano, AE ; Tondelli, M ; Meletti, S ; Kong, X-Z ; Francks, C ; Fisher, SE ; Caldairou, B ; Ryten, M ; Labate, A ; Sisodiya, SM ; Thompson, PM ; McDonald, CR ; Bernasconi, A ; Bernasconi, N ; Bernhardt, BC (OXFORD UNIV PRESS, 2022-03-25)
    Temporal lobe epilepsy, a common drug-resistant epilepsy in adults, is primarily a limbic network disorder associated with predominant unilateral hippocampal pathology. Structural MRI has provided an in vivo window into whole-brain grey matter structural alterations in temporal lobe epilepsy relative to controls, by either mapping (i) atypical inter-hemispheric asymmetry; or (ii) regional atrophy. However, similarities and differences of both atypical asymmetry and regional atrophy measures have not been systematically investigated. Here, we addressed this gap using the multisite ENIGMA-Epilepsy dataset comprising MRI brain morphological measures in 732 temporal lobe epilepsy patients and 1418 healthy controls. We compared spatial distributions of grey matter asymmetry and atrophy in temporal lobe epilepsy, contextualized their topographies relative to spatial gradients in cortical microstructure and functional connectivity calculated using 207 healthy controls obtained from Human Connectome Project and an independent dataset containing 23 temporal lobe epilepsy patients and 53 healthy controls and examined clinical associations using machine learning. We identified a marked divergence in the spatial distribution of atypical inter-hemispheric asymmetry and regional atrophy mapping. The former revealed a temporo-limbic disease signature while the latter showed diffuse and bilateral patterns. Our findings were robust across individual sites and patients. Cortical atrophy was significantly correlated with disease duration and age at seizure onset, while degrees of asymmetry did not show a significant relationship to these clinical variables. Our findings highlight that the mapping of atypical inter-hemispheric asymmetry and regional atrophy tap into two complementary aspects of temporal lobe epilepsy-related pathology, with the former revealing primary substrates in ipsilateral limbic circuits and the latter capturing bilateral disease effects. These findings refine our notion of the neuropathology of temporal lobe epilepsy and may inform future discovery and validation of complementary MRI biomarkers in temporal lobe epilepsy.
  • Item
  • Item
    No Preview Available
    Healthy Life-Year Costs of Treatment Speed From Arrival to Endovascular Thrombectomy in Patients With Ischemic Stroke A Meta-analysis of Individual Patient Data From 7 Randomized Clinical Trials
    Almekhlafi, MA ; Goyal, M ; Dippel, DWJ ; Majoie, CBLM ; Campbell, BCV ; Muir, KW ; Demchuk, AM ; Bracard, S ; Guillemin, F ; Jovin, TG ; Mitchell, P ; White, P ; Hill, MD ; Brown, S ; Saver, JL (AMER MEDICAL ASSOC, 2021-05-03)
    Importance: The benefits of endovascular thrombectomy (EVT) are time dependent. Prior studies may have underestimated the time-benefit association because time of onset is imprecisely known. Objective: To assess the lifetime outcomes associated with speed of endovascular thrombectomy in patients with acute ischemic stroke due to large-vessel occlusion (LVO). Data Sources: PubMed was searched for randomized clinical trials of stent retriever thrombectomy devices vs medical therapy in patients with anterior circulation LVO within 12 hours of last known well time, and for which a peer-reviewed, complete primary results article was published by August 1, 2020. Study Selection: All randomized clinical trials of stent retriever thrombectomy devices vs medical therapy in patients with anterior circulation LVO within 12 hours of last known well time were included. Data Extraction/Synthesis: Patient-level data regarding presenting clinical and imaging features and functional outcomes were pooled from the 7 retrieved randomized clinical trials of stent retriever thrombectomy devices (entirely or predominantly) vs medical therapy. All 7 identified trials published in a peer-reviewed journal (by August 1, 2020) contributed data. Detailed time metrics were collected including last known well-to-door (LKWTD) time; last known well/onset-to-puncture (LKWTP) time; last known well-to-reperfusion (LKWR) time; door-to-puncture (DTP) time; and door-to-reperfusion (DTR) time. Main Outcomes and Measures: Change in healthy life-years measured as disability-adjusted life-years (DALYs). DALYs were calculated as the sum of years of life lost (YLL) owing to premature mortality and years of healthy life lost because of disability (YLD). Disability weights were assigned using the utility-weighted modified Rankin Scale. Age-specific life expectancies without stroke were calculated from 2017 US National Vital Statistics. Results: Among the 781 EVT-treated patients, 406 (52.0%) were early-treated (LKWTP ≤4 hours) and 375 (48.0%) were late-treated (LKWTP >4-12 hours). In early-treated patients, LKWTD was 188 minutes (interquartile range, 151.3-214.8 minutes) and DTP 105 minutes (interquartile range, 76-135 minutes). Among the 298 of 380 (78.4%) patients with substantial reperfusion, median DTR time was 145.0 minutes (interquartile range, 111.5-185.5 minutes). Care process delays were associated with worse clinical outcomes in LKW-to-intervention intervals in early-treated patients and in door-to-intervention intervals in early-treated and late-treated patients, and not associated with LKWTD intervals, eg, in early-treated patients, for each 10-minute delay, healthy life-years lost were DTP 1.8 months vs LKWTD 0.0 months; P < .001. Considering granular time increments, the amount of healthy life-time lost associated with each 1 second of delay was DTP 2.2 hours and DTR 2.4 hours. Conclusions and Relevance: In this study, care delays were associated with loss of healthy life-years in patients with acute ischemic stroke treated with EVT, particularly in the postarrival time period. The finding that every 1 second of delay was associated with loss of 2.2 hours of healthy life may encourage continuous quality improvement in door-to-treatment times.
  • Item
    Thumbnail Image
    Effect of a comprehensive deep-learning model on the accuracy of chest x-ray interpretation by radiologists: a retrospective, multireader multicase study.
    Seah, JCY ; Tang, CHM ; Buchlak, QD ; Holt, XG ; Wardman, JB ; Aimoldin, A ; Esmaili, N ; Ahmad, H ; Pham, H ; Lambert, JF ; Hachey, B ; Hogg, SJF ; Johnston, BP ; Bennett, C ; Oakden-Rayner, L ; Brotchie, P ; Jones, CM (Elsevier BV, 2021-08)
    BACKGROUND: Chest x-rays are widely used in clinical practice; however, interpretation can be hindered by human error and a lack of experienced thoracic radiologists. Deep learning has the potential to improve the accuracy of chest x-ray interpretation. We therefore aimed to assess the accuracy of radiologists with and without the assistance of a deep-learning model. METHODS: In this retrospective study, a deep-learning model was trained on 821 681 images (284 649 patients) from five data sets from Australia, Europe, and the USA. 2568 enriched chest x-ray cases from adult patients (≥16 years) who had at least one frontal chest x-ray were included in the test dataset; cases were representative of inpatient, outpatient, and emergency settings. 20 radiologists reviewed cases with and without the assistance of the deep-learning model with a 3-month washout period. We assessed the change in accuracy of chest x-ray interpretation across 127 clinical findings when the deep-learning model was used as a decision support by calculating area under the receiver operating characteristic curve (AUC) for each radiologist with and without the deep-learning model. We also compared AUCs for the model alone with those of unassisted radiologists. If the lower bound of the adjusted 95% CI of the difference in AUC between the model and the unassisted radiologists was more than -0·05, the model was considered to be non-inferior for that finding. If the lower bound exceeded 0, the model was considered to be superior. FINDINGS: Unassisted radiologists had a macroaveraged AUC of 0·713 (95% CI 0·645-0·785) across the 127 clinical findings, compared with 0·808 (0·763-0·839) when assisted by the model. The deep-learning model statistically significantly improved the classification accuracy of radiologists for 102 (80%) of 127 clinical findings, was statistically non-inferior for 19 (15%) findings, and no findings showed a decrease in accuracy when radiologists used the deep-learning model. Unassisted radiologists had a macroaveraged mean AUC of 0·713 (0·645-0·785) across all findings, compared with 0·957 (0·954-0·959) for the model alone. Model classification alone was significantly more accurate than unassisted radiologists for 117 (94%) of 124 clinical findings predicted by the model and was non-inferior to unassisted radiologists for all other clinical findings. INTERPRETATION: This study shows the potential of a comprehensive deep-learning model to improve chest x-ray interpretation across a large breadth of clinical practice. FUNDING: Annalise.ai.
  • Item
    Thumbnail Image
    A Thalamo-Centric Neural Signature for Restructuring Negative Self-Beliefs
    Steward, T ; Kung, P-H ; Davey, C ; Moffat, B ; Glarin, R ; Jamieson, A ; Felmingham, K ; Harrison, B (ELSEVIER SCIENCE INC, 2022-05-01)
    Negative self-beliefs are a core feature of psychopathology. Despite this, we have a limited understanding of the brain mechanisms by which negative self-beliefs are cognitively restructured. Using a novel paradigm, we had participants use Socratic questioning techniques to restructure negative beliefs during ultra-high resolution 7-Tesla functional magnetic resonance imaging (UHF 7 T fMRI) scanning. Cognitive restructuring elicited prominent activation in a fronto-striato-thalamic circuit, including the mediodorsal thalamus (MD), a group of deep subcortical nuclei believed to synchronize and integrate prefrontal cortex activity, but which has seldom been directly examined with fMRI due to its small size. Increased activity was also identified in the medial prefrontal cortex (MPFC), a region consistently activated by internally focused mental processing, as well as in lateral prefrontal regions associated with regulating emotional reactivity. Using Dynamic Causal Modelling (DCM), evidence was found to support the MD as having a strong excitatory effect on the activity of regions within the broader network mediating cognitive restructuring. Moreover, the degree to which participants modulated MPFC-to-MD effective connectivity during cognitive restructuring predicted their individual tendency to engage in repetitive negative thinking. Our findings represent a major shift from a cortico-centric framework of cognition and provide important mechanistic insights into how the MD facilitates key processes in cognitive interventions for common psychiatric disorders. In addition to relaying integrative information across basal ganglia and the cortex, we propose a multifaceted role for the MD whose broad excitatory pathways act to increase synchrony between cortical regions to sustain complex mental representations, including the self.
  • Item
    Thumbnail Image
    Endovascular Therapy Versus Medical Therapy for Acute Stroke Attributable to Isolated Cervical Internal Carotid Artery Occlusion Without Intracranial Large Vessel Occlusion
    Waters, MJ ; McMullan, P ; Mitchell, PJ ; Kleinig, TJ ; Churilov, L ; Scroop, R ; Dowling, RJ ; Bush, SJ ; Nguyen, M ; Yan, B (Ovid Technologies (Wolters Kluwer Health), 2022-03)
    Background The optimal treatment for acute stroke attributable to isolated cervical internal carotid artery occlusion without intracranial target is unclear. The purpose of our study was to examine whether endovascular therapy for acute stroke attributable to isolated cervical internal carotid artery occlusion was associated with improved clinical outcome. Methods We identified patients from 2 comprehensive stroke centers during the period January 2009 to December 2019, with acute ischemic stroke attributable to cervical internal carotid artery occlusion without an intracranial occlusion. We categorized patients into 2 groups: endovascular therapy and medical therapy. Clinical outcome (modified Rankin scale score at 90 days poststroke) was compared between the 2 groups. Results Seventy‐three patients were included (26 women [36%]; median age, 69 [interquartile range (IQR), 60–80] years; median National Institutes of Health Stroke Scale score, 11 [IQR, 5–16]). Of these, 40 patients received endovascular therapy, and 33 patients were managed with medical therapy alone. The endovascular therapy group had a significantly higher median National Institutes of Health Stroke Scale score on presentation (13 versus 3; P <0.0001). Rates of thrombolysis were also significantly higher in the endovascular group (50% versus 15%; P =0.002). There were no other significant differences in baseline characteristics between the 2 groups. Good clinical outcome (modified Rankin scale score 0–2 at 90 days or no decline in modified Rankin scale score from baseline at 90 days) was seen in 73% of the endovascular therapy group compared with the 61% of the medical management group (odds ratio [OR] for good outcome, 1.7 [95% CI, 0.64–4.6]), despite the large discrepancy in baseline stroke severity. When restricted to patients with presenting National Institutes of Health Stroke Scale score ≥6, endovascular therapy was associated with higher rates of good clinical outcome (66% versus 18%; OR for good outcome, 9.0 [95% CI, 1.65–49.0]). Conclusions Endovascular therapy in isolated cervical internal carotid artery occlusion may be associated with improved outcome when compared with medical therapy. However, the significant differences in baseline characteristics between the groups limit interpretation. Randomized controlled trials are necessary.
  • Item
    Thumbnail Image
    Machine learning approaches for imaging-based prognostication of the outcome of surgery for mesial temporal lobe epilepsy
    Sinclair, B ; Cahill, V ; Seah, J ; Kitchen, A ; Vivash, LE ; Chen, Z ; Malpas, CB ; O'Shea, MF ; Desmond, PM ; Hicks, RJ ; Morokoff, AP ; King, JA ; Fabinyi, GC ; Kaye, AH ; Kwan, P ; Berkovic, SF ; Law, M ; O'Brien, TJ (WILEY, 2022-03-25)
    OBJECTIVES: Around 30% of patients undergoing surgical resection for drug-resistant mesial temporal lobe epilepsy (MTLE) do not obtain seizure freedom. Success of anterior temporal lobe resection (ATLR) critically depends on the careful selection of surgical candidates, aiming at optimizing seizure freedom while minimizing postoperative morbidity. Structural MRI and FDG-PET neuroimaging are routinely used in presurgical assessment and guide the decision to proceed to surgery. In this study, we evaluate the potential of machine learning techniques applied to standard presurgical MRI and PET imaging features to provide enhanced prognostic value relative to current practice. METHODS: Eighty two patients with drug resistant MTLE were scanned with FDG-PET pre-surgery and T1-weighted MRI pre- and postsurgery. From these images the following features of interest were derived: volume of temporal lobe (TL) hypometabolism, % of extratemporal hypometabolism, presence of contralateral TL hypometabolism, presence of hippocampal sclerosis, laterality of seizure onset volume of tissue resected and % of temporal lobe hypometabolism resected. These measures were used as predictor variables in logistic regression, support vector machines, random forests and artificial neural networks. RESULTS: In the study cohort, 24 of 82 (28.3%) who underwent an ATLR for drug-resistant MTLE did not achieve Engel Class I (i.e., free of disabling seizures) outcome at a minimum of 2 years of postoperative follow-up. We found that machine learning approaches were able to predict up to 73% of the 24 ATLR surgical patients who did not achieve a Class I outcome, at the expense of incorrect prediction for up to 31% of patients who did achieve a Class I outcome. Overall accuracies ranged from 70% to 80%, with an area under the receiver operating characteristic curve (AUC) of .75-.81. We additionally found that information regarding overall extent of both total and significantly hypometabolic tissue resected was crucial to predictive performance, with AUC dropping to .59-.62 using presurgical information alone. Incorporating the laterality of seizure onset and the choice of machine learning algorithm did not significantly change predictive performance. SIGNIFICANCE: Collectively, these results indicate that "acceptable" to "good" patient-specific prognostication for drug-resistant MTLE surgery is feasible with machine learning approaches utilizing commonly collected imaging modalities, but that information on the surgical resection region is critical for optimal prognostication.