Shehbaz Ansari

Dr. Shehbaz Ansari

orcid

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
13 July 2023 in Diagnostics
Verified
Journal Article
Open access
PDF file
FVFarzan Vahedifard
Farzan Vahedifard
HAHua Asher Ai
Hua Asher Ai
MSMark P. Supanich
Mark P. Supanich
KMKranthi Kumar Marathu
Kranthi Kumar Marathu
Xuchu LiuXuchu Liu
Xuchu Liu
SBSharon Byrd
Sharon Byrd
In this study, we developed an automated workflow using a deep learning model (DL) to measure the lateral ventricle linearly in fetal brain MRI, which are subsequently classified into normal or ventriculomegaly, defined as a diameter wider than 10 mm at the level of the thalamus and choroid plexus. To accomplish this, we first trained a UNet-based deep learning model to segment the brain of a fetus into seven different tissue categories using a public dataset (FeTA 2022) consisting of fetal T2-weighted images. Then, an automatic workflow was developed to perform lateral ventricle measurement at the level of the thalamus and choroid plexus. The test dataset included 22 cases of normal and abnormal T2-weighted fetal brain MRIs. Measurements performed by our AI model were compared with manual measurements performed by a general radiologist and a neuroradiologist. The AI model correctly classified 95% of fetal brain MRI cases into normal or ventriculomegaly. It could measure the lateral ventricle diameter in 95% of cases with less than a 1.7 mm error. The average difference between measurements was 0.90 mm in AI vs. general radiologists and 0.82 mm in AI vs. neuroradiologists, which are comparable to the difference between the two radiologists, 0.51 mm. In addition, the AI model also enabled the researchers to create 3D-reconstructed images, which better represent real anatomy than 2D images. When a manual measurement is performed, it could also provide both the right and left ventricles in just one cut, instead of two. The measurement difference between the general radiologist and the algorithm (p = 0.9827), and between the neuroradiologist and the algorithm (p = 0.2378), was not statistically significant. In contrast, the difference between general radiologists vs. neuroradiologists was statistically significant (p = 0.0043). To the best of our knowledge, this is the first study that performs 2D linear measurement of ventriculomegaly with a 3D model based on an artificial intelligence approach. The paper presents a step-by-step approach for designing an AI model based on several radiological criteria. Overall, this study showed that AI can automatically calculate the lateral ventricle in fetal brain MRIs and accurately classify them as abnormal or normal.

1 September 2024 in RadioGraphics
Verified
Journal Article
Closed access
YAYashant Aswani
Yashant Aswani
APAditi Patel
Aditi Patel
XZXin Zhan
Xin Zhan
SAShehbaz Ansari
Shehbaz Ansari
LMLeonardo Gomes Marcelino
Leonardo Gomes Marcelino
SBShweta Bhatt
Shweta Bhatt
Erdheim-Chester disease (ECD) is a rare, multisystemic, inflammatory, non-Langerhans cell histiocytic neoplasm. The discovery of recurrent and somatic mutations in the mitogen-activated protein kinase signaling pathway, most commonly BRAFV600E, has led to a reclassification of ECD from an inflammatory disorder to a neoplastic process. It is now included in the revised 2016 World Health Organization classification of hematopoietic tumors and in the Langerhans group in the revised 2016 Histiocytosis Classification of the Histiocyte Society. When symptomatic, ECD most commonly manifests with bone pain and fatigue. Also, neurologic manifestations, central diabetes insipidus, exophthalmos, and periorbital xanthelasma-like lesions are frequently encountered. Pathologic findings may vary depending on the site of biopsy and may display a spectrum of features. Thus, due to the diverse clinical presentation and variable histologic findings, imaging can often show the first sign of the disease. Radiologic findings are, however, interpreted in conjunction with clinical and histologic findings to establish the diagnosis of ECD. From providing classic findings that facilitate diagnosis to helping radiologists determine the extent of disease and predicting a prognosis, the role of radiology in ECD has evolved with the understanding of the disease itself. Insights into the molecular pathogenesis and the development of targeted therapeutic agents along with approval of vemurafenib and cobimetinib have necessitated revision of the guidelines for the management of ECD. The authors discuss various radiologic findings of ECD and differential diagnoses by using an organ system-based approach and briefly describe the revised consensus recommendations for evaluation, diagnosis, and treatment based on the International Medical Symposia on ECD from a radiologist's perspective. ©RSNA, 2024 Supplemental material is available for this article. The full digital presentation is available online.

18 August 2023 in Cancers
Verified
Journal Article
Open access
PDF file
GHGrayson W. Hooper
Grayson W. Hooper
SAShehbaz Ansari
Shehbaz Ansari
JJJason Johnson
Jason Johnson
DGdaniel ginat
daniel ginat
Imaging is essential for evaluating patients with glioblastoma. Traditionally a multimodality undertaking, CT, including CT cerebral blood profusion, PET/CT with traditional fluorine-18 fluorodeoxyglucose (18F-FDG), and MRI have been the mainstays for diagnosis and post-therapeutic assessment. However, recent advances in these modalities, in league with the emerging fields of radiomics and theranostics, may prove helpful in improving diagnostic accuracy and treating the disease.

1 May 2025 in RadioGraphics
Verified
Journal Article
Closed access
SAShehbaz Ansari
Shehbaz Ansari
ISIrene Dixe de Oliveira Santo
Irene Dixe de Oliveira Santo
PAPouria Yazdian Anari
Pouria Yazdian Anari
MBMohamed Badawy
Mohamed Badawy
JFJonathan A. Flug
Jonathan A. Flug

1 October 2024 in RadioGraphics
Verified
Journal Article
Closed access
SAShehbaz Ansari
Shehbaz Ansari
MJMiral D Jhaveri
Miral D Jhaveri
PSPokhraj Prakashchandra Suthar
Pokhraj Prakashchandra Suthar
JDJeffrey DeSanto
Jeffrey DeSanto
SGSanthosh Gaddikeri
Santhosh Gaddikeri
SVSurjith Vattoth
Surjith Vattoth

27 January 2025 in Journal of Computer Assisted Tomography
Verified
Journal Article
Open access
AAAhmad Amer
Ahmad Amer
SAShehbaz Ansari
Shehbaz Ansari
AKApollo Bs Krayyem
Apollo Bs Krayyem
SKSuprateek Kundu
Suprateek Kundu
SKSwapnil Khose
Swapnil Khose
JJJason Johnson
Jason Johnson
Background: Treatment-related changes may occur due to radiation and temozolomide in glioblastoma and can mimic tumor progression on conventional MRI. DCE-MRI enables quantification of the extent of blood-brain barrier (BBB) disruption, providing information about areas of suspicious postcontrast T1 enhancement. We compared DCE-MRI processing methods for distinguishing true disease progression from pseudoprogression in high-grade gliomas (HGGs). Methods: We identified 110 patients with HGG treated with surgery and chemoradiation who underwent DCE-MRI to further interrogate areas of new/increasing enhancement. All patients had confirmatory surgery/biopsy with pathology-confirmed progression or pseudoprogression. Scans were performed at 3T and analyzed using nordicICE. The MCA, SSS, and Parker models are three standardized processing methodologies used to create ktrans maps, a parameter that quantifies BBB permeability. Three equal regions of interest were placed at sites of peak contrast enhancement within each lesion. Data from each method was processed for mean and maximum ktrans. We conducted several rounds of analysis and finalized a strategy on penalized support vector machines based on engineered features with bootstrap sampling. Results: The Parker method was significant for ktrans maximum in the combined pathology and clinical as well as the pathology-only data sets. MCA and SSS did not perform well under the SVM classifier for pathology only. For clinical follow-up subjects, the Parker method yielded statistically significant results for maximum and mean ktrans. Conclusions: The Parker method was effective in distinguishing PD and PsP for pathology and clinical data sets. MCA and SSS techniques were effective for the clinical data set.