FDG-PET/CT and rest myocardial perfusion imaging to predict high-degree atrioventricular block recovery in cardiac sarcoidosis.

BACKGROUNDS: High-degree atrioventricular block (AVB) recovery in CS has been shown to be highly variable despite immunosuppressive treatment, with no reliable tool available to predict odds of reversibility. This study sought to evaluate the potential of combined fluorodeoxyglucose (FDG) positron emission tomography (PET)/computed tomography (CT) and resting myocardial perfusion imaging (rMPI) to predict reversibility of newly diagnosed high-grade AVB in cardiac sarcoidosis (CS). METHODS: We performed a single-center, retrospective analysis of patients with CS presenting with high-grade AVB who underwent combined FDG-PET/CT and rMPI. The 2016 JCS and the 2014 HRS diagnostic criteria were used for the diagnosis of CS. Patients with a history of coronary artery disease or prior immunosuppressive treatment were excluded. Patients were divided into AVB recovery and non-recovery subgroups. CS disease staging was based on FDG-PET and rMPI findings: (Stage 0) normal FDG-PET and rMPI (Stage 1) positive FDG-PET and normal rMPI (Stage 2) positive FDG-PET with perfusion deficits on rMPI (Stage 3) normal FDG-PET with perfusion deficits on rMPI. RESULTS: Twenty-seven patients, including 13 demonstrating AVB recovery, were identified. Eleven out of fourteen (78.6%) patients presenting with stage 1 CS demonstrated AVB recovery. Stage 1 CS was significantly more present in the recovery group compared to the non-recovery group (84.6% vs 21.4%, P = .002). Eleven presented with stage 2 CS, with only 2 (18.2%) recovering AV nodal conduction. Stage 2 CS presented more frequently in the non-recovery group (64.3% vs 15.4%, P = .020). CONCLUSIONS: Combined FDG-PET and rMPI employed to stage CS disease presenting with high-degree AVB appears to have good performance for predicting likelihood of recovery.

[18F]FDG-PET CT for the evaluation of native valve endocarditis.

BACKGROUND: We examined the use of [18F]FDG-PET/CT for the diagnosis of native valve endocarditis (NVE). METHODS: PET/CT images in patients with suspected NVE were retrospectively reviewed independently by two experienced physicians blinded to all clinical information. The gold standard consisted of surgical findings, when available, or the modified Duke criteria. RESULTS: Fifty four subjects were included, 31 (57%) with a diagnosis of NVE. [18F]FDG-PET/CT correctly identified 21/31 (67.7%) subjects, yielding a sensitivity and specificity of 68% (95% CI 49-83%) and 100% (95% CI 85-100%), respectively. The sensitivity and specificity of the modified Duke criteria were 48% and 74%, respectively. Positive and negative predictive values of PET were 100% (95% CI 84-100%) and 70% (95% CI 51-84%), respectively. Modifying the Duke criteria to include [18F]FDG-PET positivity as a major criterion increased sensitivity to 77% without affecting specificity and led to the correct reclassification of 8/18 (44.4%) subjects from Possible IE to Definite IE. CONCLUSION: The addition of a positive [18F]FDG-PET/CT as a major criterion in the modified Duke Criteria improved performance of the criteria for the diagnosis of NVE, particularly in those subjects with Possible IE.

Comparison of 18F-sodium fluoride positron emission tomography imaging and 99mTc-pyrophosphate in cardiac amyloidosis.

BACKGROUND: Pyrophosphate (PYP) scintigraphy provides high diagnostic accuracy for the detection of transthyretin (ATTR) cardiac amyloidosis (CA). There has recently been emerging interest in using 18F-sodium fluoride (NaF) for this application, yet its sensitivity has never been directly compared to that of PYP, the current molecular gold standard METHODS: Twelve subjects with ATTR-CA and 5 controls referred for PYP-SPECT were prospectively enrolled. 18F-NaF PET/CT scans were performed at 1 and 3 hours. Qualitative and quantitative analyses of the images were performed, and the sensitivity of 18F-NaF PET/CT and PYP-SPECT were compared RESULTS: Visual interpretation of NaF PET/CT yielded a sensitivity of 0.25 (95% CI 0.089 to 0.53) for the detection of ATTR-CA, which is significantly inferior to that of PYP-SPECT/CT (100%, P = .016). Visual interpretation at 3 hours yielded a similar sensitivity of 0.30 (95% CI 0.11 to 0.60, P = 1.00). There were no false-positive NaF PET studies. Mean target-to-background ratio (TBRmean) at 1h did not differ significantly (P = .21) in ATTR-CA subjects (0.83 ± 0.15) compared to controls (0.72 ± 0.15). Receiver operating characteristic (ROC) analysis yielded an area under the curve (AUC) of 0.69 ± 0.16 (95% CI 0.37 to 1.00, P = .23). CONCLUSION: With qualitative and quantitative analyses, sensitivity of NaF PET/CT is significantly inferior to that of PYP-SPECT for the diagnosis of ATTR-CA.

Improving Detection of CAD and Prognosis with PET/CT Quantitative Absolute Myocardial Blood Flow Measurements.

PURPOSE OF REVIEW: The purpose of this review is to provide an overview of the role of PET MPI in the detection of CAD, focussing on the added value of MBF for diagnosis and prognostication. RECENT FINDINGS: Positron emission tomography (PET) myocardial perfusion imaging (MPI) is increasingly used for the risk stratification of patients with suspected or established coronary artery disease (CAD). PET MPI provides accurate and reproducible non-invasive quantification of myocardial blood flow (MBF) at rest and during hyperemia, providing incremental information over conventional myocardial perfusion alone. Inclusion of MBF in PET MPI interpretation improves both its sensitivity and specificity. Moreover, quantitative MBF measurements have repeatedly been shown to offer incremental and independent prognostic information over conventional clinical markers in a broad range of conditions, including in CAD. Quantitative MBF measurement is now an established and powerful tool enabling accurate risk stratification and guiding patients' management. The role of PET MPI and flow quantification in cardiac allograft vasculopathy (CAV), which represents a particular form of CAD, will also be reviewed.

Examining the sensitivity of 18F-NaF PET for the imaging of cardiac amyloidosis.

BACKGROUND: Conventional nuclear imaging with bone-seeking radiopharmaceuticals has been shown to be a sensitive test for the detection of transthyretin cardiac amyloidosis (ATTR); however, to date, few data exist on the utility of 18F-sodium fluoride (NaF) positron emission tomography (PET) in subjects with cardiac amyloidosis (CA). METHODS: Myocardial perfusion imaging and cardiac 18F-NaF PET/CT of 7 subjects with ATTR, four with light-chain CA (AL), and four controls were retrospectively reviewed. Qualitative interpretation and quantitative analyses with average left ventricular standardized uptake values (SUVmean) and target-to-background ratios (TBRmean) were performed. RESULTS: Average TBRmean was significantly increased in subjects with ATTR (0.98 ± 0.09) compared to AL (0.85 ± 0.08, P = .026) and CTL (0.82 ± 0.07, P = .020), while SUVmean was not (P = .14). Receiver-operator characteristic (ROC) analysis yielded an area under the curve (AUC) of 0.91, with a sensitivity/specificity of 75%/100% for TBRmean using a cutoff value of 0.89 for the diagnosis of ATTR. Qualitative interpretation resulted in a sensitivity/specificity of 57%/100% for ATTR. CONCLUSIONS: While 18F-NaF PET/CT demonstrates good diagnostic accuracy for ATTR, particularly when using quantitative analysis, the low TBRmean values observed in ATTR indicate poor myocardial signal. 18F-NaF PET/CT is not yet ready for clinical use in CA until further comparison studies are performed with 99mTc-DPD/PYP.

Incremental Value of FDG-PET in the Evaluation of Cardiac Masses.

PURPOSE OF REVIEW: This study aims to review the various roles and evidence underlying the use of fluorodeoxyglucose (FDG) PET/CT and PET/MR for the assessment of cardiac masses. RECENT FINDINGS: The role of FDG-PET for the evaluation of cardiac masses continues to evolve. Studies have shown that FDG-PET is particularly well-suited for differentiating malignant from benign cardiac lesions based on their metabolic activity. Furthermore, FDG-PET is uniquely positioned to investigate patients with cardiac mass as most malignant cardiac lesions are metastasis. Finally, FDG-PET enables staging of patients with primary malignant cardiac tumor, identification of potential biopsy site, and planning of radiotherapy. FDG-PET is a complementary tool for the evaluation of patients with cardiac mass and can help differentiate benign from malignant lesions, as well as provide whole-body staging.

Bone Mineral Densitometry Reporting: Pearls and Pitfalls.

Dual-energy X-ray absorptiometry (DXA) is the method of choice for assessing bone mineral density (BMD). Unfortunately, the performance and interpretation of DXA can be challenging and errors are common. In fact, it has been reported that up to 90% of BMD reports contain at least 1 error. Errors can be the result of technique or interpretative in nature or both and can result in inappropriate diagnosis and management. In this article, we review the various types of pitfalls frequently encountered by physicians interpreting DXA studies. Being aware of these pitfalls will help readers recognize and avoid them when encountered in clinical practice.

CRAX: A simple cardiovascular risk assessment tool to predict risk of acute myocardial infarction or death.

BACKGROUND: Determining the risk of cardiovascular events is essential to optimize patient management. METHODS AND RESULTS: 5842 individuals underwent SPECT myocardial perfusion imaging (MPI) with 4.4 ± 1.2 years of follow-up. Models (the CRAX tool) were derived to predict the cumulative risk of death and acute myocardial infarction (AMI) at 1, 3, and 5 years using clinical and MPI variables. Predictors of AMI and death included age, number of hospitalizations in the 3 years preceding MPI, and left ventricular ejection fraction (LVEF). Additional predictors of death were the use of pharmacological stress, and global stress total perfusion deficit (sTPD), while transient ischemic dilation (TID), and ischemic total perfusion deficit (iTPD) change were predictive of AMI. CRAX predictions were significantly (P < .001) more accurate than clinical variables or MPI results alone, resulting in a significant net reclassification improvement (NRI, 7.5% for AMI, 14.5% death) compared to clinical variables alone. Accuracy for predicting major adverse cardiac events (MACE, comprising all-cause death, AMI, unstable angina, late revascularization) was comparable to that of AMI or death. CONCLUSIONS: CRAX is a risk assessment tool that predicts the risk of AMI, death, or MACE, and improves prediction compared to clinical variables or MPI results alone.

Tc-99m-pyrophosphate scintigraphy for the diagnosis of ATTR cardiac amyloidosis: Comparison of quantitative and semi-quantitative approaches.

BACKGROUND: ATTR cardiac amyloidosis (CA) can be diagnosed with Tc-99m-PYP scintigraphy. There are two recommended interpretative approaches: the quantitative heart-to-contralateral lung ratio (H/CL) at 1 hour and the semi-quantitative visual system at 3 hours. This study's aim was to compare both approaches and to apply the semi-quantitative method at 1 hour. METHODS: Tc-99m-PYP scans of 122 consecutive subjects were reviewed using both approaches. On 1 hour planar images, regions of interest were drawn over the heart and contralateral chest to determine H/CL. Myocardial uptake was graded on 1 and 3 hour SPECT images according to the semi-quantitative method. Concordance was examined using kappa statistics. RESULTS: 31, 10, and 81 studies were positive, negative, and equivocal, respectively, for ATTR-CA using the H/CL approach. Using the grading system, 35, 77, and 10 scans were positive, negative, and equivocal, respectively. The quantitative approach led to a significantly higher proportion of equivocal studies compared to the semi-quantitative approach (P < .0001). These approaches yielded discordant results in 2 subjects; biopsy results were concordant with SPECT grade. 1 and 3 hour SPECT grades provided concordant result in 99% of cases. CONCLUSIONS: The H/CL approach resulted in a high proportion of equivocal studies. Using SPECT imaging, the semi-quantitative approach minimized this proportion and showed high concordance at 1 and 3 hours.

Factors Associated With Bone Density Monitoring While on Antiosteoporosis Treatment in Routine Clinical Practice: A Registry-Based Cohort Study.

The role for bone mineral density (BMD) monitoring while on antiosteoporosis therapy remains controversial. The current study used population-based registries to identify factors associated with BMD monitoring in women within 5 yr of receiving antiosteoporosis treatment vs treatment without monitoring in routine clinical practice. The analytical dataset consisted of women age 40 yr and older at baseline receiving antiosteoporosis therapy: 6877 with BMD monitoring (mean interval 3.2 yr) and 6747 without BMD monitoring. There was a significant negative secular trend in BMD monitoring during the study period (p < 0.001). Multivariable logistic regression demonstrated that parental hip fracture, glucocorticoid and aromatase inhibitor use, and lower baseline BMD were independently and positively associated with BMD monitoring. Individuals with increasing age, greater body mass index, smoking, rheumatoid arthritis, later calendar year, diabetes, rural residency, lower income, and greater comorbidity score were less likely to undergo monitoring. A shorter monitoring interval (<23 mo) was strongly associated with glucocorticoid and aromatase inhibitor use. In conclusion, our study identifies factors associated with BMD monitoring over 5 yr in patients receiving antiosteoporosis therapy.

18F-Flurodeoxyglucose positron emission tomography with computed tomography (FDG PET/CT) findings in children with encephalitis and comparison to conventional imaging.

PURPOSE: FDG PET/CT is emerging as a new tool for the evaluation of acute encephalitis (AE). However, to date, there are no exclusively pediatric studies on the use of FDG PET for suspected AE. The objective of this study was to compare qualitative and quantitative brain PET to conventional brain imaging in a cohort of children, and to identify patterns of metabolic abnormalities characteristic of AE. METHODS: This retrospective study included 34 children imaged with PET/CT, CT and magnetic resonance imaging (MRI). The positivity rate of all three imaging modalities was measured. Besides visual assessment, quantification of relative regional brain metabolism (RRBM) was performed and compared to a database of normal pediatric brains. RESULTS: Fourteen subjects had a clinical diagnosis of autoimmune encephalitis (AIE) or encephalitis of unknown origin (EX), six of anti-N-methyl-D-aspartate receptor (anti-NMDAr) encephalitis, three of Hashimoto's encephalopathy, three of neurolupus and eight had other subtypes of encephalitis. Quantitative PET was abnormal in 100% of cases, visually assessed PET in 94.1% of subjects, MRI in 41.2% and CT in 6.9%. RRBM quantification demonstrated multiple hyper and hypo metabolic cortical regions in 82.3% of subjects, exclusively hypermetabolic abnormalities in 3%, and exclusively hypometabolic abnormalities in 14.7%. The basal ganglia were hypermetabolic in 26.5% of cases on visual assessment and in 58.8% of subjects using quantification. CONCLUSION: In our pediatric population FDG PET was more sensitive than conventional imaging for the detection of AE, and basal ganglia hypermetabolism was frequently encountered.

Quantification of PET Myocardial Blood Flow.

PURPOSE OF REVIEW: The aim of this review is to provide an update on quantification of myocardial blood flow (MBF) with positron emission tomography (PET) imaging. Technical and clinical aspects of flow quantification with PET are reviewed. RECENT FINDINGS: The diagnostic and prognostic values of myocardial flow quantification have been established in numerous studies and in various populations. MBF quantification has also shown itself to be particularly useful in the assessment of coronary microvascular dysfunction and in evaluation of cardiac allograft vasculopathy. Overall, myocardial flow reserve (MFR) and hyperemic MBF can lead to improved risk stratification by providing information complementary to that of other markers of disease severity, such as fractional flow reserve. Flow quantification enhances MPI's ability to detect both significant epicardial disease and microvascular dysfunction. With recent technological and methodological advances, flow quantification with PET is no longer restricted to cyclotron-equipped academic centers.

The utility and limitations of using trabecular bone score with FRAX.

PURPOSE OF REVIEW: Trabecular bone score (TBS) is a texture index derived from the lumbar spine dual-energy X-ray absorptiometry which can assess skeletal quality and provide information about fracture risk independent of bone mineral density (BMD). TBS is useful in assessing osteoporotic fracture risk, with lower TBS values associated with increased fracture risk. In this article, we review the current state of TBS, including its utility and limitations in the assessment and management of osteoporosis, with particular emphasis on the recent literature. RECENT FINDINGS: Ten-year fracture risk assessment using the FRAX tool can be improved through the use of a TBS adjustment. The use of TBS-adjusted FRAX can change management in a modest but significant number of patients, particularly in those close to an intervention threshold. Change in lumbar spine TBS for patients undergoing antiresorptive treatment is not a useful indicator of antifracture effect. SUMMARY: Lumbar spine TBS provides information complementary to conventional BMD, and has been shown to be clinically useful for enhancing fracture risk prediction.

Reproducibility of radioactive iodine uptake (RAIU) measurements.

BACKGROUND: Measurement of radioactive iodine uptake (RAIU) is an important aspect of the assessment and treatment of patients with hyperthyroidism. Its uncertainty affects how much of a true change in RAIU can be detected as well as appropriateness of the therapy dosage upon which it is based. In this study, a method of estimating the reproducibility and least significant change (LSC) values for RAIU measurements, and the implications of the values observed are discussed, with emphasis on application to quality assurance initiatives. METHODS: We prospectively studied 36 consecutive patients referred for RAIU measurements. Twenty-four hours after oral administration of 370 kBq of 131 I-NaI in capsule form, RAIU measurements were obtained in duplicate using a thyroid probe uptake system. Assessment of reproducibility was performed using root-mean-square standard deviation. RESULTS: Average difference between duplicated RAIU measurements in our study cohort was -0.1 ± 1.6% and ranged from -4.8% to 3.1%. Reproducibility of probe-based RAIU measurement was calculated to be 1.1% and 95% LSC was 3.2%. CONCLUSION: In our clinic, probe-based RAIU is a reproducible and relatively precise measurement. Using the method we have outlined, each institution can perform reproducibility assessment and compute the LSC of RAIU measurements based on its own staff, iodine isotope, equipment, protocols, and patient population. These values are useful in the assessment of serial change in RAIU, and as more experience is accumulated, can serve as benchmarks to be used in quality assurance initiatives.

Flashy lungs and sarcoidosis: not always a sign of disease activity.

This is the case of a 68-year-old man with known cardiac sarcoidosis undergoing treatment with methotrexate who presented with new onset of dyspnea and lipothymia. FDG-PET/CT revealed pathological uptake within lung parenchyma which resolved following discontinuation of methotrexate, compatible with methotrexate-induced pneumonitis. This is the first case of methotrexate-induced pneumonitis documented by FDG-PET/CT.

Clinical impact of unsuccessful subcutaneous administration of octreotide LAR instead of intramuscular administration in patients with metastatic gastroenteropancreatic neuroendocrine tumors.

Octreotide LAR is a long-acting somatostatin analogue (SSA) used in the management of metastatic gastroenteropancreatic neuroendocrine tumors (GEP NETs). It requires intramuscular (IM) injection. Missed IM injections cause subcutaneous nodules (SCNs) on radiologic images. We reviewed the rates of SCNs in a real-world cohort of GEP NETs receiving octreotide LAR and explored treatment outcomes. Patients commencing octreotide LAR between August 5, 2010 and March 8, 2018 at a single cancer center in Canada were identified from pharmacy records. Patients were included if they had a computed tomography (CT) scan performed at the time of progression and a preceding CT with pelvis included to enable assessment for the presence of nodules. Fisher's exact test was used to examine predictors of SCNs, and Kaplan-Meier curves summarized differences in progression free (PFS) and overall survival (OS) that were compared with log-rank tests. Of 243 patients receiving octreotide LAR, 45 had all required CT images available for central review. SCNs were found in 20/45 (44%) of patients on the last scan showing stable disease before progression and were numerically but not statistically more likely in females (OR: 2.36, 95% CI: 0.66-8.29, p = .23). There was an increased risk of SCNs in patients with a skin-to-muscle distance >38 mm (the length of an octreotide LAR needle) on CT (OR: 5.09, 95% CI: 1.39-16.6, p = .018) and a trend toward increased risk in obese patients (OR: 5.71, 95% CI: 1.26-23.4, p = .061). PFS (HR: 1.01, 95% CI: 0.56-1.78, p = .98) and OS (HR: 0.86, 95% CI: 0.41-1.8, p = .70) was similar between those with/without SCNs. In conclusion, almost half of patients receiving octreotide LAR had SCNs; however, missed administration of SSA did not appear to result in worse survival in this small study. Factors such as sex, younger age skin-to-muscle distance, and obesity may affect SCN development and should be considered when choosing an SSA.

177Lu-PSMA-617 in Metastatic Castration-Resistant Prostate Cancer: A Review of the Evidence and Implications for Canadian Clinical Practice.

Prostate-specific membrane antigen (PSMA) is highly expressed in prostate cancer and a therapeutic target. Lutetium-177 (177Lu)-PSMA-617 is the first radioligand therapy to be approved in Canada for use in patients with metastatic castration-resistant prostate cancer (mCRPC). As this treatment represents a new therapeutic class, guidance regarding how to integrate it into clinical practice is needed. This article aims to review the evidence from prospective phase 2 and 3 clinical trials and meta-analyses of observational studies on the use of 177Lu-PSMA-617 in prostate cancer and discuss how Canadian clinicians might best apply these data in practice. The selection of appropriate patients, the practicalities of treatment administration, including necessary facilities for treatment procedures, the assessment of treatment response, and the management of adverse events are considered. Survival benefits were observed in clinical trials of 177Lu-PSMA-617 in patients with progressive, PSMA-positive mCRPC who were pretreated with androgen receptor pathway inhibitors and taxanes, as well as in taxane-naïve patients. However, the results of ongoing trials are awaited to clarify questions regarding the optimal sequencing of 177Lu-PSMA-617 with other therapies, as well as the implications of predictive biomarkers, personalized dosimetry, and combinations with other therapies.

Semi-supervised learning towards automated segmentation of PET images with limited annotations: application to lymphoma patients.

Manual segmentation poses a time-consuming challenge for disease quantification, therapy evaluation, treatment planning, and outcome prediction. Convolutional neural networks (CNNs) hold promise in accurately identifying tumor locations and boundaries in PET scans. However, a major hurdle is the extensive amount of supervised and annotated data necessary for training. To overcome this limitation, this study explores semi-supervised approaches utilizing unlabeled data, specifically focusing on PET images of diffuse large B-cell lymphoma (DLBCL) and primary mediastinal large B-cell lymphoma (PMBCL) obtained from two centers. We considered 2-[18F]FDG PET images of 292 patients PMBCL (n = 104) and DLBCL (n = 188) (n = 232 for training and validation, and n = 60 for external testing). We harnessed classical wisdom embedded in traditional segmentation methods, such as the fuzzy clustering loss function (FCM), to tailor the training strategy for a 3D U-Net model, incorporating both supervised and unsupervised learning approaches. Various supervision levels were explored, including fully supervised methods with labeled FCM and unified focal/Dice loss, unsupervised methods with robust FCM (RFCM) and Mumford-Shah (MS) loss, and semi-supervised methods combining FCM with supervised Dice loss (MS + Dice) or labeled FCM (RFCM + FCM). The unified loss function yielded higher Dice scores (0.73 ± 0.11; 95% CI 0.67-0.8) than Dice loss (p value < 0.01). Among the semi-supervised approaches, RFCM + αFCM (α = 0.3) showed the best performance, with Dice score of 0.68 ± 0.10 (95% CI 0.45-0.77), outperforming MS + αDice for any supervision level (any α) (p < 0.01). Another semi-supervised approach with MS + αDice (α = 0.2) achieved Dice score of 0.59 ± 0.09 (95% CI 0.44-0.76) surpassing other supervision levels (p < 0.01). Given the time-consuming nature of manual delineations and the inconsistencies they may introduce, semi-supervised approaches hold promise for automating medical imaging segmentation workflows.

Optimizing stress-only myocardial perfusion imaging: a clinical prediction model to improve patient selection.

BACKGROUND: Stress-only single photon emission computed tomography myocardial perfusion imaging (MPI) offers numerous advantages in terms of improved workflow, cost and radiation reduction but is currently not widely utilized due to challenges in selecting appropriate patients for this technique. METHODS: Data from 5959 individuals were used to derive (N = 4018) and validate (N = 1941) a binomial logistic regression model to predict normal stress MPI studies (stress total perfusion deficit  < 4%, ejection fraction ≥ 50%). Model performance was analyzed using receiver operator characteristic curves. A simplified point-scoring system was developed and its impact on imaging workflow was assessed. RESULTS: Significant predictors of abnormal vs. normal stress MPI included male sex, age > 65 years, cardiomyopathy, congestive heart failure, myocardial infarction, angina, and pharmacological stress. The final model and simplified scoring system were associated with areas under the curve of 0.81 (95% CI 0.79-0.83) and 0.80 (95% CI 0.79-0.82) in the validation group, respectively. Use of the scoring system was estimated to result in a decrease of 56.5% in the number of non-contributory imaging studies acquired with minimal patient rescheduling. CONCLUSION: A prediction tool derived from simple clinical information can identify candidates for stress-only MPI studies with a beneficial impact on departmental workflow.

Predicting maximum temperatures over India 10-days ahead using machine learning models.

In the months of March-June, India experiences high daytime temperatures (Tmax), which sometimes lead to heatwave-like conditions over India. In this study, 10 different machine learning models are evaluated for their ability to predict the daily Tmax anomalies 10 days ahead in the months of March-June. Several model experiments were carried out to identify an optimal model to predict daily Tmax anomalies over India. The results indicate that the AdaBoost regressor with Multi-layer Perceptron as the base estimator is an optimal model to predict the Tmax anomalies over India in the months of March-June. The optimal model predictions are benchmarked against 10-day persistence predictions and the predictions from the Climate Forecast System (CFS) reforecast. The results indicate that the machine learning model skill is higher than persistence and comparable to CFS reforecast 10-day predictions in April and May. In March and June, the machine learning models have low skill scores and perform no better than persistence. These results indicate that the machine learning models are promising tools to predict the surface air maximum temperature anomalies over India in April and May and can complement predictions from more sophisticated numerical models.