Diagnostic performance of an artificial intelligence-driven cardiac-structured reporting system for myocardial perfusion SPECT imaging.

OBJECTIVES: To describe and validate an artificial intelligence (AI)-driven structured reporting system by direct comparison of automatically generated reports to results from actual clinical reports generated by nuclear cardiology experts. BACKGROUND: Quantitative parameters extracted from myocardial perfusion imaging (MPI) studies are used by our AI reporting system to generate automatically a guideline-compliant structured report (sR). METHOD: A new nonparametric approach generates distribution functions of rest and stress, perfusion, and thickening, for each of 17 left ventricle segments that are then transformed to certainty factors (CFs) that a segment is hypoperfused, ischemic. These CFs are then input to our set of heuristic rules used to reach diagnostic findings and impressions propagated into a sR referred as an AI-driven structured report (AIsR). The diagnostic accuracy of the AIsR for detecting coronary artery disease (CAD) and ischemia was tested in 1,000 patients who had undergone rest/stress SPECT MPI. RESULTS: At the high-specificity (SP) level, in a subset of 100 patients, there were no statistical differences in the agreements between the AIsr, and nine experts' impressions of CAD (P = .33) or ischemia (P = .37). This high-SP level also yielded the highest accuracy across global and regional results in the 1,000 patients. These accuracies were statistically significantly better than the other two levels [sensitivity (SN)/SP tradeoff, high SN] across all comparisons. CONCLUSIONS: This AI reporting system automatically generates a structured natural language report with a diagnostic performance comparable to those of experts.

Clinical decision support systems in myocardial perfusion imaging.

Diagnostic imaging is becoming more complicated, physicians are also required to master an ever-expanding knowledge base and take into account an ever increasing amount of patient-specific clinical information while the time available to master this knowledge base, assemble the relevant clinical data, and apply it to specific tasks is steadily shrinking. Compounding these problems, there is an ever increasing number of aging "Baby Boomers" who are becoming patients coupled with a declining number of cardiac diagnosticians experienced in interpreting these studies. Hence, it is crucial that decision support tools be developed and implemented to assist physicians in interpreting studies at a faster rate and at the highest level of up-to-date expertise. Such tools will minimize subjectivity and intra- and inter-observer variation in image interpretation, help achieve a standardized high level of performance, and reduce healthcare costs. Presently, there are many decision support systems and approaches being developed and implemented to provide greater automation and to further objectify and standardize analysis, display, integration, interpretation, and reporting of myocardial perfusion SPECT and PET studies. This review focuses on these systems and approaches.

Hypertension-mediated organ damage in masked hypertension.

OBJECTIVES: Masked hypertension - a blood pressure (BP) phenotype characterized by a clinic BP in the normal range but elevated BP outside the office - is associated with early hypertension-mediated organ damage. This study examined early target organ manifestations of masked hypertension diagnosed by home (HBPM) and ambulatory (ABPM) BP monitoring. METHODS: Left ventricular (LV) structure and diastolic function measured by echocardiography, microalbuminuria, and coronary artery calcification were evaluated in 420 patients with high clinic BP (SBP 120-150 mmHg or DBP 80-95 mmHg). Evidence of hypertension-mediated organ damage was compared in patients with sustained normotension, masked hypertension, and sustained hypertension based on measurements by HBPM, daytime ABPM, and 24-h ABPM. RESULTS: The 420 participants averaged 48 (12) [mean (SD)] years of age; the average clinic BP was 130 (13)/81 (8) mmHg. In individuals with masked hypertension diagnosed by HBPM, indexed LV mass, relative wall thickness, and e' and E/e' (indices of LV relaxation), were generally intermediate between values observed in normotensives and sustained hypertensive patients, and were significantly greater in masked hypertension than normotensives. Similar trends were observed when masked hypertension was diagnosed by ABPM but a diagnosis of masked hypertension was not as reliably associated with LV remodeling or impaired LV relaxation in comparison to normotensives. There were trends towards greater likelihoods of detectable urinary microalbumin and coronary calcification in masked hypertension than in normotensives. CONCLUSION: These results support previous studies demonstrating early hypertension-mediated organ damage in patients with masked hypertension, and suggest that HBPM may be superior to ABPM in identifying patients with masked hypertension who have early LV remodeling and diastolic LV dysfunction.

Coronary Artery Calcifications and Cardiac Risk After Radiation Therapy for Stage III Lung Cancer.

PURPOSE: Heart dose and heart disease increase the risk for cardiac toxicity associated with radiation therapy. We hypothesized that computed tomography (CT) coronary calcifications are associated with cardiac toxicity and may help ascertain baseline heart disease. METHODS AND MATERIALS: We analyzed the cumulative incidence of cardiac events in patients with stage III non-small cell lung cancer receiving median 74 Gy on prospective dose-escalation trials. Events were defined as symptomatic effusion, pericarditis, unstable angina, infarction, significant arrhythmia, and/or heart failure. Coronary calcifications were delineated on simulation CTs using radiation software program (130 HU threshold). Calcifications were defined as "none," "low," and "high," with median volume dividing low and high. RESULTS: Of 109 patients, 26 had cardiac events at median 26 months (range, 1-84 months) after radiation therapy. Median follow-up in surviving patients was 8.8 years (range, 2.3-17.3). On simulation CTs, 64 patients (59%) had coronary calcifications with median volume 0.2 cm3 (range, 0.01-8.3). Only 16 patients (15%) had baseline coronary artery disease. Cardiac events occurred in 7% (3 of 45), 29% (9 of 31), and 42% (14 of 33) of patients with no, low, and high calcifications, respectively. Calcification burden was associated with cardiac toxicity on univariate (low vs none: hazard ratio [HR] 5.0, P = .015; high vs none: HR 8.1, P < .001) and multivariate analyses (low vs none: HR 7.0, P = .005, high vs none: HR 10.6, P < .001, heart mean dose: HR 1.1/Gy, P < .001). Four-year competing risk-adjusted event rates for no, low, and high calcifications were 4%, 23%, and 34%, respectively. CONCLUSIONS: The presence of coronary calcifications is a cardiac risk factor that can identify high-risk patients for medical referral and help guide clinicians before potentially cardiotoxic cancer treatments.

Scintographic evidence of severe myocardial hypoperfusion in a patient with left anterior descending coronary artery bridging--case report and review of the literature.

Myocardial bridging is a congenital abnormality characterized by an intramyocardial course of a major epicardial coronary artery segment. Generally considered a benign condition, myocardial bridging has been associated with angina, acute myocardial infarction, and sudden death. Herein, we report a patient with an intramyocardial segment in the mid portion of the left anterior descending coronary artery with marked systolic compression. Single photon emission computed tomography with technetium-99m tetrofosmin done to evaluate an episode of chest pain showed a large predominantly fixed perfusion defect in the mid to apical anterior wall with partial reversibility. The patient's chest pain did not recur and repeat single photon emission computed tomography imaging 14 days later with rest-redistribution thallium-201 showed normal myocardial perfusion. The overall clinical impression was that myocardial bridging resulted in severe transient anterior myocardial hypoperfusion. The literature on prevalence, diagnosis, use of perfusion imaging, and hemodynamic effects of myocardial bridging is reviewed.

Three-dimensional fusion of coronary arteries with myocardial perfusion distributions: clinical validation.

UNLABELLED: Clinical decisions regarding diagnosis and effective treatment of coronary artery disease frequently require integration of information from various imaging modalities, which are acquired, processed, and read at different physical locations and times. We have developed methods to integrate the information in 2 cardiac imaging studies, perfusion SPECT and coronary angiography. Three-dimensional (3D) models of the coronary artery tree created from biplane angiograms were automatically aligned with 3D models of the left ventricular epicardial surface created from perfusion SPECT. Myocardial mass at risk was used as a unique measure to validate the accuracy of the unification. METHODS: Thirty patients were injected with the perfusion agent (99m)Tc-tetrosfosmin during balloon occlusion while undergoing percutaneous transluminal coronary angioplasty for single-vessel coronary artery disease. Thus, a single, severe perfusion defect was induced by a single coronary artery occlusion of known severity and placement. The accuracy of the unification was measured by computing the overlap between physiologic area at risk, determined using SPECT perfusion quantification techniques only, and anatomic area at risk, determined using coronary artery anatomy aligned with the epicardial surface of the left ventricle. RESULTS: The unification resulted in an 80% overlap of areas at risk, and an overlap of 84% of normal areas, for all coronary artery distributions. The mass at risk measured based on the unified anatomic information correlated with the physiologically based mass at risk as y = 0.92x + 10.3 g; r = 0.76, SEE = 10.4 g. CONCLUSION: A unification algorithm for automatically registering 3D models of the epicardial surface from perfusion SPECT and 3D coronary artery trees from coronary angiography has been presented and validated in 30 patient studies.

Reproducibility of masked hypertension among adults 30 years or older.

OBJECTIVE: Masked hypertension (MH) refers to nonelevated office blood pressure (BP) with elevated out-of-office BP, but its reproducibility has not been conclusively established. We examined 1-week reproducibility of MH by home BP monitoring (HBPM) and ambulatory BP monitoring (ABPM). METHODS: We recruited 420 adults not on BP-lowering medication, with recent clinic BP between 120/80 and 149/95 mmHg. For main comparisons, participants with office average less than 140/90 mmHg were considered to have MH if awake ABPM average was 135/85 mmHg or higher; they were considered to have MH by HBPM if the average was 135/85 mmHg or higher. Percentage agreements were quantified in terms of κ. We also examined the prevalence of MH, defined as office average less than 140/90 mmHg, with a 24-h ABPM average of 130/80 mmHg or higher. We carried out sensitivity analyses using different threshold BP levels for ABPM-office pairings and HBPM-office pairings for defining MH. RESULTS: Prevalence rates of MH based on office-awake ABPM pairings were 44 and 43%, with an agreement of 71% (κ=0.40; 95% confidence interval 0.31-0.49). MH was less prevalent (15 and 17%) using HBPM-office pairings, with agreement of 82% (κ=0.30; 95% confidence interval 0.16-0.44), and more prevalent when considering the 24-h average (50 and 48%). MH was also less prevalent when more stringent diagnostic criteria were applied. Office-HBPM pairings and office-awake ABPM pairings had fair agreement on MH classification on both occasions, with κ-values of 0.36 and 0.30. CONCLUSION: MH has fair short-term reproducibility, providing further evidence that for some people, out-of-office BP is systematically higher than that measured in the office setting.

Image-guided optimization of the ECG trace in cardiac MRI.

Improper electrocardiogram (ECG) lead placement resulting in suboptimal gating may lead to reduced image quality in cardiac magnetic resonance imaging (CMR). A patientspecific systematic technique for rapid optimization of lead placement may improve CMR image quality. A rapid 3 dimensional image of the thorax was used to guide the realignment of ECG leads relative to the cardiac axis of the patient in forty consecutive adult patients. Using our novel approach and consensus reading of pre- and post-correction ECG traces, seventy-three percent of patients had a qualitative improvement in their ECG tracings, and no patient had a decrease in quality of their ECG tracing following the correction technique. Statistically significant improvement was observed independent of gender, body mass index, and cardiac rhythm. This technique provides an efficient option to improve the quality of the ECG tracing in patients who have a poor quality ECG with standard techniques.

Superposición en tres dimensiones de las imágenes de perfusión miocárdica y de la coronariografía / Unified three-dimensional images of myocardial perfusion and coronary angiography

Introducción y objetivos. Integrar la información anatómica y funcional en pacientes con cardiopatía isquémica es una tarea habitual en la práctica diaria del cardiólogo. El objetivo de este trabajo es presentar una metodología de unificación tridimensional de la información anatómica relativa a las coronarias epicárdicas, proveniente de la coronariografía, con la información fisiológica de perfusión tisular procedente de la tomogammagrafía de perfusión miocárdica. Métodos. Se seleccionaron 3 pacientes programados, por criterios clínicos, para revascularización coronaria percutánea y con enfermedad de un solo vaso (descendente anterior, coronaria derecha y circunfleja). Las imágenes coronariográficas biplanares se obtuvieron antes y después de la dilatación. Durante la oclusión coronaria se administró una dosis de 99mTc-tetrofosmina y, una vez finalizada la dilatación, se obtuvieron las imágenes de perfusión miocárdica correspondientes a la oclusión. La tomogammagrafía de control en reposo se repitió dos días después. Mediante una metodología propia se generaron los contornos epicárdicos de las imágenes de perfusión miocárdica y sobre ellos se superpuso el árbol coronario tridimensional proveniente de la coronariografía. Resultados. Se logró una correcta reconstrucción tridimensional del contorno epicárdico y del árbol coronario completo en los 3 pacientes. La imagen unificada en tres dimensiones presentó una excelente concordancia entre la extensión de los defectos de perfusión y la distribución anatómica del vaso ocluido. Conclusiones. La superposición tridimensional de las imágenes de perfusión miocárdica y de la coronariografía es técnicamente posible. Ello permite integrar la información anatómica y funcional de cara a facilitar la toma de decisiones por parte del cardiólogo y mejorar el manejo del paciente coronario (AU)