Faculty Bibliography
Searched for:
in-biosketch:true
person:leem69
Total Results:
30
Association of partial T2-FLAIR mismatch sign and isocitrate dehydrogenase mutation in WHO grade 4 gliomas: results from the ReSPOND consortium
PURPOSE/OBJECTIVE:While the T2-FLAIR mismatch sign is highly specific for isocitrate dehydrogenase (IDH)-mutant, 1p/19q-noncodeleted astrocytomas among lower-grade gliomas, its utility in WHO grade 4 gliomas is not well-studied. We derived the partial T2-FLAIR mismatch sign as an imaging biomarker for IDH mutation in WHO grade 4 gliomas. METHODS:Preoperative MRI scans of adult WHO grade 4 glioma patients (n = 2165) from the multi-institutional ReSPOND (Radiomics Signatures for PrecisiON Diagnostics) consortium were analyzed. Diagnostic performance of the partial T2-FLAIR mismatch sign was evaluated. Subset analyses were performed to assess associations of imaging markers with overall survival (OS). RESULTS:One hundred twenty-one (5.6%) of 2165 grade 4 gliomas were IDH-mutant. Partial T2-FLAIR mismatch was present in 40 (1.8%) cases, 32 of which were IDH-mutant, yielding 26.4% sensitivity, 99.6% specificity, 80.0% positive predictive value, and 95.8% negative predictive value. Multivariate logistic regression demonstrated IDH mutation was significantly associated with partial T2-FLAIR mismatch (odds ratio [OR] 5.715, 95% CI [1.896, 17.221], p = 0.002), younger age (OR 0.911 [0.895, 0.927], p < 0.001), tumor centered in frontal lobe (OR 3.842, [2.361, 6.251], p < 0.001), absence of multicentricity (OR 0.173, [0.049, 0.612], p = 0.007), and presence of cystic (OR 6.596, [3.023, 14.391], p < 0.001) or non-enhancing solid components (OR 6.069, [3.371, 10.928], p < 0.001). Multivariate Cox analysis demonstrated cystic components (p = 0.024) and non-enhancing solid components (p = 0.003) were associated with longer OS, while older age (p < 0.001), frontal lobe center (p = 0.008), multifocality (p < 0.001), and multicentricity (p < 0.001) were associated with shorter OS. CONCLUSION/CONCLUSIONS:Partial T2-FLAIR mismatch sign is highly specific for IDH mutation in WHO grade 4 gliomas.
PMID: 37468750
ISSN: 1432-1920
CID: 5535892
Distinguishing Brain Metastasis Progression From Radiation Effects After Stereotactic Radiosurgery Using Longitudinal GRASP Dynamic Contrast-Enhanced MRI
BACKGROUND:Differentiating brain metastasis progression from radiation effects or radiation necrosis (RN) remains challenging. Golden-angle radial sparse parallel (GRASP) dynamic contrast-enhanced MRI provides high spatial and temporal resolution to analyze tissue enhancement, which may differ between tumor progression (TP) and RN. OBJECTIVE:To investigate the utility of longitudinal GRASP MRI in distinguishing TP from RN after gamma knife stereotactic radiosurgery (SRS). METHODS:We retrospectively evaluated 48 patients with brain metastasis managed with SRS at our institution from 2013 to 2020 who had GRASP MRI before and at least once after SRS. TP (n = 16) was pathologically confirmed. RN (n = 16) was diagnosed on either resected tissue without evidence of tumor or on lesion resolution on follow-up. As a reference, we included a separate group of patients with non-small-cell lung cancer that showed favorable response with tumor control and without RN on subsequent imaging (n = 16). Mean contrast washin and washout slopes normalized to the superior sagittal sinus were compared between groups. Receiver operating characteristic analysis was performed to determine diagnostic performance. RESULTS:After SRS, progression showed a significantly steeper washin slope than RN on all 3 follow-up scans (scan 1: 0.29 ± 0.16 vs 0.18 ± 0.08, P = .021; scan 2: 0.35 ± 0.19 vs 0.18 ± 0.09, P = .004; scan 3: 0.32 ± 0.12 vs 0.17 ± 0.07, P = .002). No significant differences were found in the post-SRS washout slope. Post-SRS washin slope differentiated progression and RN with an area under the curve (AUC) of 0.74, a sensitivity of 75%, and a specificity of 69% on scan 1; an AUC of 0.85, a sensitivity of 92%, and a specificity of 69% on scan 2; and an AUC of 0.87, a sensitivity of 63%, and a specificity of 100% on scan 3. CONCLUSION:Longitudinal GRASP MRI may help to differentiate metastasis progression from RN.
PMID: 36700674
ISSN: 1524-4040
CID: 5419632
Federated learning enables big data for rare cancer boundary detection
Although machine learning (ML) has shown promise across disciplines, out-of-sample generalizability is concerning. This is currently addressed by sharing multi-site data, but such centralization is challenging/infeasible to scale due to various limitations. Federated ML (FL) provides an alternative paradigm for accurate and generalizable ML, by only sharing numerical model updates. Here we present the largest FL study to-date, involving data from 71 sites across 6 continents, to generate an automatic tumor boundary detector for the rare disease of glioblastoma, reporting the largest such dataset in the literature (n = 6, 314). We demonstrate a 33% delineation improvement for the surgically targetable tumor, and 23% for the complete tumor extent, over a publicly trained model. We anticipate our study to: 1) enable more healthcare studies informed by large diverse data, ensuring meaningful results for rare diseases and underrepresented populations, 2) facilitate further analyses for glioblastoma by releasing our consensus model, and 3) demonstrate the FL effectiveness at such scale and task-complexity as a paradigm shift for multi-site collaborations, alleviating the need for data-sharing.
PMCID:9722782
PMID: 36470898
ISSN: 2041-1723
CID: 5381682
Preoperative differentiation of hypophysitis and pituitary adenomas using a novel clinicoradiologic scoring system
PURPOSE/OBJECTIVE:Hypophysitis can clinically and radiologically mimic other nonfunctioning masses of the sella turcica, complicating preoperative diagnosis. While sellar masses may be treated surgically, hypophysitis is often treated medically, so differentiating between them facilitates optimal management. The objective of our study was to develop a scoring system for the preoperative diagnosis of hypophysitis. METHODS:A thorough literature review identified published hypophysitis cases, which were compared to a retrospective group of non-functioning pituitary adenomas (NFA) from our institution. A preoperative hypophysitis scoring system was developed and internally validated. RESULTS:Fifty-six pathologically confirmed hypophysitis cases were identified in the literature. After excluding individual cases with missing values, 18 hypophysitis cases were compared to an age- and sex-matched control group of 56 NFAs. Diabetes insipidus (DI) (p < 0.001), infundibular thickening (p < 0.001), absence of cavernous sinus invasion (CSI) (p < 0.001), relation to pregnancy (p = 0.002), and absence of visual symptoms (p = 0.007) were significantly associated with hypophysitis. Stepwise logistic regression identified DI and infundibular thickening as positive predictors of hypophysitis. CSI and visual symptoms were negative predictors. A 6-point hypophysitis-risk scoring system was derived: + 2 for DI, + 2 for absence of CSI, + 1 for infundibular thickening, + 1 for absence of visual symptoms. Scores ≥ 3 supported a diagnosis of hypophysitis (AUC 0.96, sensitivity 100%, specificity 75%). The scoring system identified 100% of hypophysitis cases at our institution with an estimated 24.7% false-positive rate. CONCLUSIONS:The proposed scoring system may aid preoperative diagnosis of hypophysitis, preventing unnecessary surgery in these patients.
PMID: 35622211
ISSN: 1573-7403
CID: 5229072
Deep multi-task learning and random forest for series classification by pulse sequence type and orientation
PURPOSE/OBJECTIVE:Increasingly complex MRI studies and variable series naming conventions reveal limitations of rule-based image routing, especially in health systems with multiple scanners and sites. Accurate methods to identify series based on image content would aid post-processing and PACS viewing. Recent deep/machine learning efforts classify 5-8 basic brain MR sequences. We present an ensemble model combining a convolutional neural network and a random forest classifier to differentiate 25 brain sequences and image orientation. METHODS:Series were grouped by descriptions into 25 sequences and 4 orientations. Dataset A, obtained from our institution, was divided into training (16,828 studies; 48,512 series; 112,028 images), validation (4746 studies; 16,612 series; 26,222 images) and test sets (6348 studies; 58,705 series; 3,314,018 images). Dataset B, obtained from a separate hospital, was used for out-of-domain external validation (1252 studies; 2150 series; 234,944 images). We developed an ensemble model combining a 2D convolutional neural network with a custom multi-task learning architecture and random forest classifier trained on DICOM metadata to classify sequence and orientation by series. RESULTS:The neural network, random forest, and ensemble achieved 95%, 97%, and 98% overall sequence accuracy on dataset A, and 98%, 99%, and 99% accuracy on dataset B, respectively. All models achieved > 99% orientation accuracy on both datasets. CONCLUSION/CONCLUSIONS:The ensemble model for series identification accommodates the complexity of brain MRI studies in state-of-the-art clinical practice. Expanding on previous work demonstrating proof-of-concept, our approach is more comprehensive with greater sequence diversity and orientation classification.
PMID: 35906437
ISSN: 1432-1920
CID: 5277082
A No-Math Primer on the Principles of Machine Learning for Radiologists
Machine learning is becoming increasingly important in both research and clinical applications in radiology due to recent technological developments, particularly in deep learning. As these technologies are translated toward clinical practice, there is a need for radiologists and radiology trainees to understand the basic principles behind them. This primer provides an accessible introduction to the vocabulary and concepts that are central to machine learning and relevant to the radiologist.
PMID: 35339253
ISSN: 1558-5034
CID: 5190662
Tumor volume improves preoperative differentiation of prolactinomas and nonfunctioning pituitary adenomas
PURPOSE/OBJECTIVE:Both prolactinomas and nonfunctioning adenomas (NFAs) can present with hyperprolactinemia. Distinguishing them is critical because prolactinomas are effectively managed with dopamine agonists, whereas compressive NFAs are treated surgically. Current guidelines rely only on serum prolactin (PRL) levels, which are neither sensitive nor specific enough. Recent studies suggest that accounting for tumor volume may improve diagnosis. The objective of this study is to investigate the diagnostic utility of PRL, tumor volume, and imaging features in differentiating prolactinoma and NFA. METHODS:Adult patients with pathologically confirmed prolactinoma (n = 21) or NFA with hyperprolactinemia (n = 58) between 2013 and 2020 were retrospectively identified. Diagnostic performance of clinical and imaging variables was analyzed using receiver-operating characteristic curves to calculate area under the curve (AUC). RESULTS:with sensitivity of 100% and specificity of 82.76%. Binary logistic regression found that PRL was a significant positive predictor of prolactinoma diagnosis, whereas tumor volume, presence of CSI not previously defined, and T2 hyperintensity were significant negative predictors. The regression model had an AUC of 0.9915 (p < 0.0001). CONCLUSIONS:Consideration of tumor volume improves differentiation between prolactinomas and NFAs, which in turn leads to effective management.
PMID: 33966173
ISSN: 1559-0100
CID: 4878192
"The Pituitary within GRASP" - Golden-Angle Radial Sparse Parallel Dynamic MRI Technique and Applications to the Pituitary Gland
MRI is the preferred radiologic modality for evaluating the pituitary gland. An important component of pituitary MRI examinations is dynamic contrast-enhanced MRI. Compared to conventional dynamic techniques, golden-angle radial sparse parallel (GRASP) imaging offers multiple advantages, including the ability to achieve higher spatial and temporal resolution. In this narrative review, we discuss dynamic imaging of the pituitary gland, the technical fundamentals of GRASP, and applications of GRASP to the pituitary gland.
PMID: 34147165
ISSN: 1558-5034
CID: 4924692
Bilateral Retrocerebellar Arachnoid Cysts Presenting as Peripheral Vertigo: Case Report and Literature Review [Case Report]
Bilateral retrocerebellar arachnoid cysts are exceedingly rare. We report a case of a 38-year-old woman, who presented with progressive vertigo and was found to have bilateral retrocerebellar arachnoid cysts. The patient's clinical presentation was most consistent with benign positional peripheral vertigo, while the cysts were thought to be incidental findings. We review the literature on bilateral retrocerebellar arachnoid cysts and discuss their management.
PMCID:7417180
PMID: 32789077
ISSN: 2168-8184
CID: 5904222
Utility of Percentage Signal Recovery and Baseline Signal in DSC-MRI Optimized for Relative CBV Measurement for Differentiating Glioblastoma, Lymphoma, Metastasis, and Meningioma
BACKGROUND AND PURPOSE:The percentage signal recovery in non-leakage-corrected (no preload, high flip angle, intermediate TE) DSC-MR imaging is known to differ significantly for glioblastoma, metastasis, and primary CNS lymphoma. Because the percentage signal recovery is influenced by preload and pulse sequence parameters, we investigated whether the percentage signal recovery can still differentiate these common contrast-enhancing neoplasms using a DSC-MR imaging protocol designed for relative CBV accuracy (preload, intermediate flip angle, low TE). MATERIALS AND METHODS:= 13) using a protocol designed for relative CBV accuracy (a one-quarter-dose preload and single-dose bolus of gadobutrol, TR/TE = 1290/40 ms, flip angle = 60° at 1.5T). Mean percentage signal recovery, relative CBV, and normalized baseline signal intensity were compared within contrast-enhancing lesion volumes. Classification accuracy was determined by receiver operating characteristic analysis. RESULTS:Relative CBV best differentiated meningioma from glioblastoma and from metastasis with areas under the curve of 0.84 and 0.82, respectively. The percentage signal recovery best differentiated primary central nervous system lymphoma from metastasis with an area under the curve of 0.81. Relative CBV and percentage signal recovery were similar in differentiating primary central nervous system lymphoma from glioblastoma and from meningioma. Although neither relative CBV nor percentage signal recovery differentiated glioblastoma from metastasis, mean normalized baseline signal intensity achieved 86% sensitivity and 50% specificity. CONCLUSIONS:Similar to results for non-preload-based DSC-MR imaging, percentage signal recovery for one-quarter-dose preload-based, intermediate flip angle DSC-MR imaging differentiates most pair-wise comparisons of glioblastoma, metastasis, primary central nervous system lymphoma, and meningioma, except for glioblastoma versus metastasis. Differences in normalized post-preload baseline signal for glioblastoma and metastasis, reflecting a snapshot of dynamic contrast enhancement, may motivate the use of single-dose multiecho protocols permitting simultaneous quantification of DSC-MR imaging and dynamic contrast-enhanced MR imaging parameters.
PMCID:6728173
PMID: 31371360
ISSN: 1936-959x
CID: 5904212
