Parametric Cerebrovascular Reserve Images Using Acetazolamide (99m)Tc-HMPAO SPECT: A Feasibility Study of Quantitative Assessment

Basal and Acetazolamide Brain Perfusion SPECT in Internal Carotid Artery Stenosis

Internal carotid artery (ICA) stenosis including Moyamoya disease needs revascularization when hemodynamic insufficiency is validated. Vascular reserve impairment was the key to find the indication for endarterectomy/bypass surgery in the atherosclerotic ICA stenosis and to determine the indication, treatment effect, and prognosis in Moyamoya diseases. Vascular reserve was quantitatively assessed by 1-day split-dose I-123 IMP basal/acetazolamide SPECT in Japan or by Tc-99m HMPAO SPECT in other countries using qualitative or semi-quantitative method. We summarized the development of 1-day basal/ acetazolamide brain perfusion SPECT for ICA stenosis, both quantitative and qualitative methods, and their methodological issues regarding (1) acquisition protocol; (2) qualitative assessment, either visual or deep learning-based; (3) clinical use for atherosclerotic ICA steno-occlusive diseases and mostly Moyamoya diseases; and (4) their impact on the choice of treatment options. Trials to use CT perfusion or perfusion MRI using contrast materials or arterial spin labeling were briefly discussed in their endeavor to use basal studies alone to replace acetazolamide-challenge SPECT. Theoretical and practical issues imply that basal perfusion evaluation, no matter how much sophisticated, will not disclose vascular reserve. Acetazolamide rarely causes serious adverse reactions but included fatality, and now, we need to monitor patients closely in acetazolamide-challenge studies.

Deep learning-based interpretation of basal/acetazolamide brain perfusion SPECT leveraging unstructured reading reports

PURPOSE

Basal/acetazolamide brain perfusion single-photon emission computed tomography (SPECT) has been used to evaluate functional hemodynamics in patients with carotid artery stenosis. We aimed to develop a deep learning model as a support system for interpreting brain perfusion SPECT leveraging unstructured text reports.

METHODS

In total, 7345 basal/acetazolamide brain perfusion SPECT images and their text reports were retrospectively collected. A long short-term memory (LSTM) network was trained using 500 randomly selected text reports to predict manually labeled structured information, including abnormalities of basal perfusion and vascular reserve for each vascular territory. Using this trained LSTM model, we extracted structured information from the remaining 6845 text reports to develop a deep learning model for interpreting SPECT images. The model was based on a 3D convolutional neural network (CNN), and the performance was tested on the other 500 cases by measuring the area under the receiver-operating characteristic curve (AUC). We then applied the model to patients who underwent revascularization (n = 33) to compare the estimated output of the CNN model for pre- and post-revascularization SPECT and clinical outcomes.

RESULTS

The AUC of the LSTM model for extracting structured labels was 1.00 for basal perfusion and 0.99 for vascular reserve for all 9 brain regions. The AUC of the CNN model designed to identify abnormal perfusion was 0.83 for basal perfusion and 0.89 for vascular reserve. The output of the CNN model was significantly improved according to the revascularization in the target vascular territory, and its changes in brain territories were concordant with clinical outcomes.

CONCLUSION

We developed a deep learning model to support the interpretation of brain perfusion SPECT by converting unstructured text reports into structured labels. This model can be used as a support system not only to identify perfusion abnormalities but also to provide quantitative scores of abnormalities, particularly for patients who require revascularization.

Imaging Diagnosis of Transient Ischemic Attack in Clinic and Traditional Chinese Medicine

Neuroimaging plays a pivotal role in Transient Ischemic Attack (TIA). Generally, clinicians focus on the specific changes in morphology and function, but the diagnosis of TIA often depends on imaging evidence. Whereas Traditional Chinese Medicine (TCM) is concerned with the performance of clinical symptoms, they began to use imaging methods to diagnose TIA. CT and MRI are the recommended modality to diagnose TIA and image ischemic lesions. In addition, Transcranial Doppler sonography (TCD) and Digital Subtraction Angiography (DSA) are two acceptable alternatives for diagnosing TIA patients. This article elaborates the update of imaging modalities in clinic and the development of imaging modalities in TCM. Besides, multiple joint imaging technologies also will be evaluated whether enhanced diagnostic yields availably.

The Current Status of SPECT or SPECT/CT in South Korea

The first step to nuclear medicine in Korea started with introduction of the gamma camera in 1969. Although planar images with the gamma camera give important functional information, they have the limitations that result from 2-dimensional images. Single-photon emission computed tomography (SPECT) due to its 3-dimensional image acquisition is superior to earlier planar gamma imaging in image resolution and diagnostic accuracy. As demand for a hybrid functional and anatomical imaging device has increased, integrated SPECT/CT systems have been used. In Korea, SPECT/CT was for the first time installed in 2003. SPECT/CT can eliminate many possible pitfalls on SPECT-alone images, making better attenuation correction and thereby improving image quality. Therefore, SPECT/CT is clinically preferred in many hospitals in various aspects. More recently, additional SPECT/CT images taken from the region with equivocal uptake on planar images have been helpful in making precise interpretation as part of their clinical workup in postoperative thyroid cancer patients. SPECT and SPECT/CT have various advantages, but its clinical application has gradually decreased in recent few years. While some researchers investigated the myocardial blood flow with cardiac PET using F-18 FDG or N-13 ammonia, myocardial perfusion SPECT is, at present, the radionuclide imaging study of choice for the risk stratification and guiding therapy in the coronary artery disease patients in Korea. New diagnostic radiopharmaceuticals for AD have received increasing attention; nevertheless, brain SPECT will remain the most reliable modality evaluating cerebral perfusion.