Evaluation of Role of F-18 FDG Cardiac PET and Tc-99m Sestamibi Myocardial Perfusion Imaging in Assessing the Therapeutic Benefit in Patients with Coronary Artery Disease and Left Ventricular Systolic Dysfunction

(RSNA 2018, Sun Nov 25 2018 10:55AM – 11:05AM ROOM Z44)

PURPOSE

To evaluate the therapeutic benefit with revascularization and optimal medical treatment (OMT) in patients diagnosed with hibernating myocardium on myocardial perfusion imaging (MPI) using F-18 FDG cardiac PET

METHOD AND MATERIALS

59 consecutive patients (43 males, 16 females, Mean Age 60.7 ± 9.4 years) with CAD and LV systolic dysfunction who underwent myocardial viability imaging for revascularization work-up and were diagnosed with hibernating myocardium were enrolled in this study. Patients were later treated with either revascularization or OMT and were followed for a median duration of 7.7 months for assessing the therapeutic benefit. Therapeutic benefit was assessed under 3 categories (a) Improvement in functional class (b) Adverse cardiac-events and (c) Improvement in LV function and myocardial perfusion on follow-up resting 99mTc-sestamibi myocardial perfusion imaging.

RESULTS

29 patients underwent revascularization (49%) and 25 patients received OMT (42%). Five patients were lost to follow-up. Patients were matched for baseline characteristics in both treatment arms. On follow-up, significant improvement was noted in NYHA functional class and CCS angina class post-revascularization. No such improvement was noted in the OMT group. The cardiac-event rate of patients in OMT group was significantly higher than that of patients in revascularization group (36% vs. 10.3 %; p = 0.046). At 1 year of follow-up, event-free survival in revascularization group was significantly superior compared to OMT group (83.8% vs. 50.8%; p= 0.039). On follow-up resting MPI scan, mean improvement in LVEF in revascularization group was significantly higher than in OMT group (6.0% vs. 1.4%; p=0.04).

CONCLUSION

Myocardial viability imaging is a sensitive modality to identify hibernating myocardium in patients with CAD and LV dysfunction and predicting its recovery following revascularization, thereby guiding the optimal treatment strategy for these patients.

CLINICAL RELEVANCE/APPLICATION

Myocardial viability imaging should be performed prior to revascularization in patients with coronary artery disease with left-ventricular dysfunction to help predict recovery post-treatment.

Improving the Accuracy of Deep Learning Networks for Bone-Age Estimation by Incorporating Radiological Insight Guided Feature Analysis

(RSNA 2018, Wed Nov 28 2018 9:00AM – 9:10AM ROOM Z09)

PURPOSE

The Greulich-Pyle (GP) method of bone age determination primarily involves estimation of ossification of the epiphyseal centers around the radiocarpal, carpometacarpal and the proximal interphalangeal joints and the carpal bones. This radiological insight was applied to devise a novel sequential approach, involving segmentation of relevant wrist anatomy from hand X-Rays followed by deep learning, and compared it against a standard deep learning technique to assess bone age in pediatric population between 7 years to 18 years.

METHOD AND MATERIALS

Dataset containing 12,600 radiographs provided by RSNA for Bone Age Challenge is used for this work. Out of the 12,600 we used ~10,000 radiographs of children between 7 years to 18 years of age. The intensity values of the hand radiographs are standardized across dataset by histogram matching image pre-processing techniques. A pre-processing algorithm was created to crop relevant regions, i.e. proximal phalanges, metacarpals, carpals and distal ends of radius and ulna, from the hand radiographs. Finally, ~9,000 cropped images were used to train a convolutional neural network implemented in the research version of HealthSuite Insights (Philips HealthTech) to predict the bone age from the image. The remaining images (~1,000) were used for validation purposes. Additional datasets of 200 test images released by RSNA and 50 test images obtained as part of routine clinical practice (extracted from PACS and anonymised using HIPAA compliant methods) were used to calculate Mean Absolute Error (MAE). Similar MAE was calculated for the same convolutional neural network implemented in the research version of HealthSuite Insights (Philips HealthTech) trained on the images without cropping the images.

RESULTS

The performance of deep learning model trained using cropped images was found to be superior with MAE of 5.08 months compared to the model trained using the full radiographs, which had MAE of 5.51 months.

CONCLUSION

The accuracy of machine learning models for specific tasks on radiographs can be improved by training using cropped/segmented radiographs containing areas of anatomy relevant to the task.

CLINICAL RELEVANCE/APPLICATION

Automated and accurate bone age estimation has wide-ranging clinical and medicolegal applications. Our novel method combines radiologist guided feature-based analysis with deep learning to improve the accuracy.

Localization and Restaging of Carcinoma Prostate by 68Ga PSMA PET/CT in Patients with Biochemical Recurrence: A Descriptive Study

(RSNA 2018, Mon Nov 26 2018 11:25AM – 11:35AM ROOM Z44)

PURPOSE

Prostate cancer is the most common solid cancer in men. Following definitive treatment of prostate cancer by radical prostatectomy (RP) or radiotherapy, cancer recurrence is heralded by an increase in serum prostate-specific antigen (PSA) which is called biochemical recurrence. We investigate the relationship between prostate specific antigen (PSA) level and detection of suspected cancer recurrence using 68 Ga-PSMA PET/CT in patients with biochemical recurrence after radical prostatectomy (RP) or radiotherapy.

METHOD AND MATERIALS

We analyzed retrospective data of 150 men with carcinoma prostate post RP and post radiotherapy with biochemical recurrence from May 2014 to Jan 2018 by 68 Ga-PSMA PET/CT We included men with suspected recurrent prostate cancer based on an elevated post treatment PSA level. The data collected analyzed the relationship of the pre-scan PSA level to the probability of a positive scan finding for recurrent prostate cancer.

RESULTS

Our cohort included 150 men, all had adenocarcinoma of prostate, 126/150 had a previous RP and 24/150 had prior radiotherapy. The mean PSA of the RP group was 4.8 ng/mL and 22.8 ng/mL in the radiotherapy group. In the post RP cohort, the detection rate of 68 Ga-PSMA PET/ CT was 39.3% for PSA 0.2 to <0.5 ng/mL, 45.3% for PSA 0.5 to <1 ng/mL, 88.2% for PSA 1 to <2 ng/mL and 95.5% for PSA ≥2. Lymph node metastasis post RP was identified in 52% of men with suspected disease recurrence. In the post radiotherapy cohort the detection rate was 96.1 % for PSA 2 to 4 ng/mL, 99.2% for PSA 4 to 6 ng/ mL and 100% for PSA ≥6. Local recurrence after radiotherapy was present in 62 % of the cohort and 58 % had lymph node metastasis. CONCLUSION 68Ga-PSMA PET/CT provides a novel imaging modality for the detection of prostate cancer recurrence and metastasis. Suspected PSMA avid metastatic lesions are common and are identified at low post treatment PSA levels, which if detected will help direct appropriate salvage treatments. CLINICAL RELEVANCE/APPLICATION PSMA PET/CT should be considered a routine part of follow-up of treated prostate cancer patients since metastasis may present with low PSA levels leading to delay in addressing relapses.

Mapping the MAPCAs with Dual Source CT: What Do Cardiothoracic Surgeons Want to Know?

(Educational exhibit, RSNA 2018)
V Venugopal, MD, New Delhi, India; V Mahajan, MBBS; H Mahajan, MD, MBBS

TEACHING POINTS
1. Major Aortopulmonary collateral arteries are unique lesions in which the pulmonary vascular bed is multi-compartmentalized
2. Unifocalization refers to the process of changing an abnormal multi-compartment pulmonary artery circulation to a normal single compartment circulation.
3. If neither a PV or ductus is present during primary morphogenesis, the foregut source of PA persists and the native pulmonary arteries do not form normally
4. What Does the Surgeon Need to Know Before Unifocalization?
– True pulmonary artery size and arborization
– Number, origin, exact course, and destination of every collateral
– Exact position and severity of all stenoses in both true pulmonary arteries and collaterals
– For every collateral, does it intercommunicate with true pulmonary artery: “isolated supply” or “dual supply”
– Relationship of collaterals to other thoracic structures: bronchial tree, pulmonary veins, esophagus
– Post-stenotic pressure in collaterals

TABLE OF CONTENTS/OUTLINE

– Basic Physiology of MAPCA
– Principles of Surgical Management
– What does the surgeon want to know
– Imaging protocols
– Dual source CT advantages
– Pitfalls and Challenges
– Some case examples

Five Free Radiology Hacks Every Practicing Radiologist Should Know

(Educational Exhibit, RSNA 2018)
V Mahajan, MBBS, New Delhi, India; V Venugopal, MD

TEACHING POINTS

Many free tools available which can enhance radiologists’ practice from a clinical, academic and management perspective.
Five main tools discussed:

RadiAnt Dicom Viewer – a <3MB tool that can open any dicom file, query from almost any PACS and has advanced visualisation tools like MIP, MPR, 3D-Recon and PET-CT/MR fusion. Orthanc Dicom Server – Takes <1 minute to install. Full fledged PACS server that can be used to view and share cases on a mobile phone or any web browser. ITK-SNAP – Building the next great deep learning algorithm? You’ll need to segment out the relevant anatomy first – learn to do that with ITK-SNAP and you won’t need to mark each slice separately. Notepad++ - Do you download each report manually for your research? Ask your IT team for an SQL dump and you can use Notepad++ to extract the cases you want. FuzzyLookup – A Microsoft Excel plugin that you can use to match wrongly written patient names, unique ID’s and just about anything. Link your reports and images easily when mining large unorganised datasets. TABLE OF CONTENTS/OUTLINE • Day-to-day problems faced by radiologists • Concept of open source software • How to use Radiant Dicom Viewer, Orthanc Dicom Server, ITK-SNAP, FuzzyLookup and Notepad++

Amide Proton Transfer Imaging for Brain Tumors: Initial Experience and Current Status

(Educational Exhibit, RSNA 2018)
V Mahajan, MBBS, MBA; R Bhattacharjee, MENG; I Saha, PhD; H Mahajan, MD; S Panwar, MD; V Venugopal, MD

TEACHING POINTS

• Amide Proton Transfer (APT) Images contain quantitative colour maps that show protein / peptide content.
• APT imaging is able to accurately predict tumour grade in pre-surgical cases.
• APT imaging is showing promise in differentiating tumour recurrence from radiation necrosis.
• APT imaging can be used in combination with Susceptibility Weighted Imaging (SWI) based ITSS (Intra-Tumoral Susceptibility Score) to predict tumour grade when compared to Contrast-Enhanced Perfusion MRI.
• Some fundamental problems with the APT acquisition need to be addressed for it to come into day-to-day clinical practice, especially with respect to quantification and artefacts.

TABLE OF CONTENTS/OUTLINE

• Review of current brain tumour MR imaging protocols
• Mechanism behind APT imaging
• Review of current clinical literature on APT imaging for brain tumors
• APT for tumour recurrence vs radiation necrosis
• APT in glioma grading
• APT with SWI-based ITSS as a non-contrast replacement for CE-Perfusion MR
• Limitations and future directions for APT imaging in the brain

Development and Validation of Deep Learning Algorithms for Detection of Critical Findings in Head CT Scans

Sasank Chilamkurthy, Rohit Ghosh, Swetha Tanamala, Mustafa Biviji, Norbert G. Campeau, Vasantha Kumar Venugopal, Vidur Mahajan, Pooja Rao, Prashant Warier
Arxiv pre-print, March 2018
ABSTRACT
Importance: Non-contrast head CT scan is the current standard for initial imaging of patients with head trauma or stroke symptoms.
Objective: To develop and validate a set of deep learning algorithms for automated detection of following key findings from non-contrast head CT scans: intracranial hemorrhage (ICH) and its types, intraparenchymal (IPH), intraventricular (IVH), subdural (SDH), extradural (EDH) and subarachnoid (SAH) hemorrhages, calvarial fractures, midline shift and mass effect.
Design and Settings: We retrospectively collected a dataset containing 313,318 head CT scans along with their clinical reports from various centers. A part of this dataset (Qure25k dataset) was used to validate and the rest to develop algorithms. Additionally, a dataset (CQ500 dataset) was collected from different centers in two batches B1 & B2 to clinically validate the algorithms.
Main Outcomes and Measures: Original clinical radiology report and consensus of three independent radiologists were considered as gold standard for Qure25k and CQ500 datasets respectively. Area under receiver operating characteristics curve (AUC) for each finding was primarily used to evaluate the algorithms.
Results: Qure25k dataset contained 21,095 scans (mean age 43.31; 42.87% female) while batches B1 and B2 of CQ500 dataset consisted of 214 (mean age 43.40; 43.92% female) and 277 (mean age 51.70; 30.31% female) scans respectively. On Qure25k dataset, the algorithms achieved an AUC of 0.9194, 0.8977, 0.9559, 0.9161, 0.9288 and 0.9044 for detecting ICH, IPH, IVH, SDH, EDH and SAH respectively. AUCs for the same on CQ500 dataset were 0.9419, 0.9544, 0.9310, 0.9521, 0.9731 and 0.9574 respectively. For detecting calvarial fractures, midline shift and mass effect, AUCs on CQ500 dataset were 0.9244, 0.9276 and 0.8583 respectively, while AUCs on Qure25k dataset were 0.9624, 0.9697 and 0.9216 respectively.

Semi-Automated Quantitative Analysis of Cartilage Thickness and T2 Values

Rafeek T, Sandeep Panwar Jogi, Sriram Rajan, Amit Mehndiratta, Anup Singh
Accepted for presentation at ISMRM, 2018

Purpose: Osteoarthris (OA) is the degenerative disease in knee joint, associated with cartilage degradation. In case of damage, the normal rate of healing of articular cartilage is quite low, hence the early detection of OA is essential for better management.

View Article

Quantitative MR Imaging of Articular Knee Cartilage with Axial Loading during Image Acquisition

Sandeep Panwar Jogi, Rafeek T., Sriram Rajan, Vidur Mahajan, Sitikanta Roy, Anup Singh, and Amit Mehndiratta
Accepted for presentation at ISMRM, 2018

Purpose: Musculoskeletal diseases and articular disorders are one of the major health concern worldwide. Among these disorder, osteoarthritis (OA) is one of common and multifarious disorder.

View Article

Automated Seed Points Selection Based Radial-Search Segmentation Method For Sagittal and Coronal View Knee MRI Imaging

Sandeep Panwar Jogi, Rafeek T., Sriram Rajan, Krithika Rangarajan, Anup Singh, and Amit Mehndiratta
Accepted for presentation at ISMRM, 2018

Synopsis: Knee disorders are generally marked in tibio-femoral bone junction. Most of available segmentation techniques use time consuming semiautomatic approach as radial search method, in sagittal view only. However, coronal view MRI Knee images are clinically equal important. Proposed approach automates seed points selection process for the radial search method, which work equally good on both sagittal and coronal view for identification of tibio-femoral junction.

View Article