Unveiling myocardial microstructure shifts: exploring the impact of diabetes in stable CAD patients through CMR T1 mapping.

BACKGROUND: This study investigates myocardial structural changes in stable coronary artery disease (CAD) patients with type 2 diabetes (T2D) using cardiac magnetic resonance (CMR) strain and T1 mapping. METHODS: A total of 155 stable CAD patients underwent CMR examination, including left ventricular (LV) morphology and function assessment, late gadolinium enhancement (LGE), and feature tracking (CMR-FT) for LV global longitudinal, circumferential, and radial strain. T1 mapping with extracellular volume (ECV) evaluation was also performed. RESULTS: Among the enrolled patients, 67 had T2D. Diabetic patients exhibited impaired LV strain and higher ECV compared to non-diabetics. Multivariate analysis identified T2D as an independent predictor of increased ECV and decreased strain. CONCLUSIONS: CMR-based strain and T1 mapping highlighted impaired myocardial contractility, elevated ECV, and potential interstitial fibrosis in diabetic patients with stable CAD. This suggests a significant impact of diabetes on myocardial health beyond CAD, emphasizing the importance of a comprehensive assessment in these individuals. TRIAL REGISTRATION: http://www.controlled-trials.com/ISRCTN09454308.

Mitral Annular Disease at Cardiac MRI: What to Know and Look For.

Accurate evaluation of the mitral valve (MV) apparatus is essential for understanding the mechanisms of MV disease across various clinical scenarios. The mitral annulus (MA) is a complex and crucial structure that supports MV function; however, conventional imaging techniques have limitations in fully capturing the entirety of the MA. Moreover, recognizing annular changes might aid in identifying patients who may benefit from advanced cardiac imaging and interventions. Multimodality cardiovascular imaging plays a major role in the diagnosis, prognosis, and management of MV disease. Transthoracic echocardiography is the first-line modality for evaluation of the MA, but it has limitations. Cardiac MRI (CMR) has emerged as a robust imaging modality for assessing annular changes, with distinct advantages over other imaging techniques, including accurate flow and volumetric quantification and assessment of variations in the measurements and shape of the MA during the cardiac cycle. Mitral annular disjunction (MAD) is defined as atrial displacement of the hinge point of the MV annulus away from the ventricular myocardium, a condition that is now more frequently diagnosed and studied owing to recent technical advances in cardiac imaging. However, several unresolved issues regarding MAD, such as the functional significance of pathologic disjunction and how this disjunction advances in the clinical course, require further investigation. The authors review the role of CMR in the assessment of MA disease, with a focus on MAD and its functional implications in MV prolapse and mitral regurgitation. ©RSNA, 2024 Supplemental material is available for this article. See the invited commentary by Stojanovska and Fujikura in this issue.

Post-COVID-19 respiratory sequelae two years after hospitalization: an ambidirectional study.

Background: COVID-19 lung sequelae can impact the course of patient lives. We investigated the evolution of pulmonary abnormalities in post-COVID-19 patients 18-24 months after hospital discharge. Methods: A cohort of COVID-19 patients admitted to the Hospital das Clínicas da Faculdade de Medicina da USP in São Paulo, Brazil, between March and August of 2020, were followed-up 6-12 months after hospital discharge. A subset of patients with pulmonary involvement and chest computed tomography (CT) scans were eligible to participate in this second follow-up (18-24 months). Data was analyzed in an ambidirectional manner, including retrospective data from the hospitalization, and from the first follow-up (6-12 months after discharge), and compared with the prospective data collected in this new follow-up. Findings: From 348 patients eligible, 237 (68%) participated in this follow-up. Among participants, 139 (58%) patients presented ground-glass opacities and reticulations, and 80 (33%) presented fibrotic-like lesions (traction bronchiectasis and architectural distortion). Five (2%) patients improved compared to the 6-12-month assessment, but 20 (25%) of 80 presented worsening of lung abnormalities. For those with relevant assessments on both occasions, comparing the CT findings between this follow-up with the previous assessment, there was an increase in patients with architectural distortion (43 [21%] of 204 vs 57 [28%] of 204, p = 0.0093), as well as in traction bronchiectasis (55 [27%] of 204 vs 69 [34%] of 204, p = 0.0043). Patients presented a persistent functional impairment with demonstrated restrictive pattern in both follow-ups (87 [42%] of 207 vs 91 [44%] of 207, p = 0.76), as well as for the reduced diffusion capacity (88 [42%] of 208 vs 87 [42%] of 208, p = 1.0). Length of hospitalization (OR 1.04 [1.01-1.07], p = 0.0040), invasive mechanical ventilation (OR 3.11 [1.3-7.5] p = 0.011), patient's age (OR 1.03 [1.01-1.06] p = 0.0074 were consistent predictors for development of fibrotic-like lung lesions in post-COVID-19 patients. Interpretation: Post-COVID-19 lung sequelae can persist and progress after hospital discharge, suggesting airways involvement and formation of new fibrotic-like lesions, mainly in patients who were in intensive care unit (ICU). Funding: São Paulo Research Foundation (22/01769-5) and Instituto Todos pela Saúde (C1721).

Safety and possible anti-inflammatory effect of paclitaxel associated with LDL-like nanoparticles (LDE) in patients with chronic coronary artery disease: a double-blind, placebo-controlled pilot study.

Introduction: Studies in cholesterol-fed rabbits showed that anti-proliferative chemotherapeutic agents such as paclitaxel associated with solid lipid nanoparticles (LDE) have marked anti-atherosclerotic effects. In addition, association with LDE nearly abolishes paclitaxel toxicity. We investigated whether treatment with LDE-paclitaxel changes plaque progression by coronary CT angiography and is safe in patients with chronic coronary artery disease. Methods: We conducted a prospective, randomized, double-blind, placebo-controlled pilot study in patients with multi-vessel chronic coronary artery disease. Patients were randomized to receive IV infusions of LDE-paclitaxel (paclitaxel dose: 175 mg/m2 body surface) or LDE alone (placebo group), administered every 3 weeks for 18 weeks. All participants received guideline-directed medical therapy. Clinical and laboratory safety evaluations were made at baseline and every 3 weeks until the end of the study. Analysis of inflammatory biomarkers and coronary CTA was also performed at baseline and 4 weeks after treatment. Results: Forty patients aged 65.6 ± 8 years, 20 in LDE-paclitaxel and 20 in placebo group were enrolled. Among those, 58% had diabetes, 50% had myocardial infarction, and 91% were in use of statin and aspirin. Baseline demographics, risk factors, and laboratory results were not different between groups. In all patients, no clinical or laboratory toxicities were observed. From the baseline to the end of follow-up, there was a non-significant trend toward a decrease in IL-6 levels and hsCRP in the LDE-paclitaxel group (-16% and -28%, respectively), not observed in placebo. Regarding plaque progression analysis, variation in plaque parameter values was wide, and no difference between groups was observed. Conclusion: In patients with multivessel chronic coronary artery disease and optimized medical therapy, LDE-paclitaxel was safe and showed clues of potential benefits in reducing inflammatory biomarkers. Clinical Trial Registration: https://clinicaltrials.gov/study/NCT04148833, identifier (NCT04148833).

Association between obstructive coronary disease and diabetic retinopathy: Cross-sectional study of coronary angiotomography and multimodal retinal imaging.

AIMS: To investigate the association between diabetic retinopathy (DR) and coronary artery disease (CAD) using coronary angiotomography (CCTA) and multimodal retinal imaging (MMRI) with ultra-widefield retinography and optical coherence tomography angiography and structural domain. METHODS: Single-center, cross-sectional, single-blind. Patients with diabetes who had undergone CCTA underwent MMRI. Uni and multivariate analysis were used to assess the association between CAD and DR and to identify variables independently associated with DR. RESULTS: We included 171 patients, 87 CAD and 84 non-CAD. Most CAD patients were males (74 % vs 38 %, P < 0.01), insulin users (52 % vs 38 %, p < 0.01) and revascularized (64 %). They had a higher prevalence of DR (48 % vs 22 %, p = 0.01), microaneurysms (25 % vs 13 %, p = 0.04), intraretinal cysts (22 % vs 8 %, p = 0.01) and areas of reduced capillary density (46 % vs 20 %, p < 0.01). CAD patients also had lower mean vascular density (MVD) (15.7 % vs 16.5,%, p = 0.049) and foveal avascular zone (FAZ) circularity (0.64 ± 0.1 vs 0.69 ± 0.1, p = 0.04). There were significant and negative correlations between Duke coronary score and MVD (r = -0.189; p = 0.03) and FAZ circularity (r = -0,206; p = 0.02). CAD, DM duration and insulin use independently associated with DR. CONCLUSIONS: CAD patients had higher prevalence of DR and lower MVD. CAD, DM duration and insulin use were independently associated with DR.

Myocardial microstructure assessed by T1 mapping after on-pump and off-pump coronary artery bypass grafting.

Background: The correlation between the release of cardiac biomarkers after revascularization, in the absence of late gadolinium enhancement (LGE) or myocardial edema, and the development of myocardial tissue damage remains unclear. This study sought to identify whether the release of biomarkers is associated with cardiac damage by assessing myocardial microstructure on T1 mapping after on-pump (ONCAB) and off-pump coronary artery bypass grafting (OPCAB). Methods: Seventy-six patients with stable multivessel coronary artery disease (CAD) and preserved systolic ventricular function were included. T1 mapping, high-sensitive cardiac troponin I (cTnI), creatine kinase myocardial band (CK-MB) mass, and ventricular dimensions and function were measured before and after procedures. Results: Of the 76 patients, 44 underwent OPCAB, and 32 ONCAB; 52 were men (68.4%), and the mean age was 63±8.5 years. In both OPCAB and ONCAB the native T1 values were similar before and after surgeries. An increase in extracellular volume (ECV) values after the procedures was observed, due to the decrease in hematocrit levels during the second cardiac resonance. However, the lambda partition coefficient showed no significant difference after the surgeries. The median peak release of cTnI and CK-MB were higher after ONCAB than after OPCAB [3.55 (2.12-4.9) vs. 2.19 (0.69-3.4) ng/mL, P=0.009 and 28.7 (18.2-55.4) vs. 14.3 (9.3-29.2) ng/mL, P=0.009, respectively]. Left ventricular ejection fraction (LVEF) was similar in both groups before and after surgery. Conclusions: In the absence of documented myocardial infarction, T1 mapping did not identify structural tissue damage after surgical revascularization with or without cardiopulmonary bypass (CPB), despite the excessive release of cardiac biomarkers.

Abnormal release of cardiac biomarkers in the presence of myocardial oedema evaluated by cardiac magnetic resonance after uncomplicated revascularization procedures.

AIMS: To analyse the association of myocardial oedema (ME), observed as high T2 signal intensity (HT2) in cardiac magnetic resonance imaging, with the release of cardiac biomarkers, ventricular ejection, and clinical outcomes after revascularization. METHODS AND RESULTS: Patients with stable coronary artery disease with the indication for revascularization were included. Biomarker levels [troponin I (cTnI) and creatine kinase MB (CK-MB)] and T2-weighted and late gadolinium enhancement (LGE) images were obtained before and after the percutaneous or surgical revascularization procedures. The association of HT2 with the levels of biomarkers, with and without LGE, evolution of left ventricular ejection fraction (LVEF), and 5-year clinical outcomes were assessed. A total of 196 patients were divided into 2 groups: Group 1 (HT2, 40) and Group 2 (no HT2, 156). Both peak cTnI (8.9 and 1.6 ng/mL) and peak CK-MB values (44.7 and 12.1 ng/mL) were significantly higher in Group 1. Based on the presence of new LGE, patients were stratified into Groups A (no HT2/LGE, 149), B (HT2, 9), C (LGE, 7), and D (both HT2/LGE, 31). The peak cTnI and CK-MB values were 1.5 and 12.0, 5.4 and 44.7, 5.0 and 18.3, and 9.8 and 42.8 ng/mL in Groups A, B, C, and D, respectively, and were significantly different. The average LVEF decreased by 4.4% in Group 1 and increased by 2.2% in Group 2 (P = 0.057). CONCLUSION: ME after revascularization procedures was associated with increased release of cardiac necrosis biomarkers, and a trend towards a difference in LVEF, indicating a role of ME in cardiac injury after interventions.

Non-contrast transcatheter aortic valve implantation for patients with aortic stenosis and chronic kidney disease: a pilot study.

Background: Acute kidney injury (AKI) is frequently observed after transcatheter aortic valve implantation (TAVI). Of note, it is associated with a threefold increase in all-cause and cardiac death. We propose a new non-contrast strategy for evaluating and performing the TAVI procedure that can be especially valuable for patients with aortic stenosis (AS) and chronic kidney disease (CKD) to prevent AKI. Methods: Patients with severe symptomatic AS and CKD stage ≥3a were evaluated for TAVI using four non-contrast imaging modalities for procedural planning: transesophageal echocardiogram (TEE), cardiac magnetic resonance, multidetector computed tomography (MDCT), and aortoiliac CO2 angiography. Patients underwent transfemoral (TF) TAVI using the self-expandable Evolut R/Pro, and the procedures were guided by fluoroscopy and TEE. Contrast MDCT and contrast injection at certain checkpoints during the procedure were used in a blinded fashion to guarantee patient safety. Results: A total of 25 patients underwent TF-TAVI with the zero-contrast technique. The mean age was 79.9 ± 6.1 years, 72% in NYHA class III/IV, with a mean STS-PROM of 3.0% ± 1.5%, and creatinine clearance of 49 ± 7 ml/min. The self-expandable Evolut R and Pro were implanted in 80% and 20% of patients, respectively. In 36% of the cases, the transcatheter heart valve (THV) chosen was one size larger than the one by contrast MDCT, but none of these cases presented adverse events. Device success and the combined safety endpoint (at 30 days) both achieved 92%. Pacemaker implantation was needed in 17%. Conclusion: This pilot study demonstrated that the zero-contrast technique for procedural planning and THV implantation was feasible and safe and might become the preferable strategy for a significant population of CKD patients undergoing TAVR. Future studies with a larger number of patients are still needed to confirm such interesting findings.

Myocardial tissue microstructure with and without stress-induced ischemia assessed by T1 mapping in patients with stable coronary artery disease.

BACKGROUND: Stress-induced myocardial ischemia seems not to be associated with cardiovascular events. However, its effects on myocardial tissue characteristics remain under debate. Thus, we sought to assess whether documented stress-induced ischemia is associated with changes in myocardial microstructure evaluated by magnetic resonance native T1 map and extracellular volume fraction (ECV). METHODS: This is a single-center, analysis of the previously published MASS V Trial. Multivessel patients with a formal indication for myocardial revascularization and with documented stress-induced ischemia were included in this study. Native T1 and ECV values evaluated by cardiac magnetic resonance imaging of ischemic and nonischemic myocardial segments at rest and after stress were compared. Myocardial ischemia was detected by either nuclear scintigraphy or stress magnetic cardiac resonance protocol. RESULTS: Between May 2012 and March 2014, 326 prospective patients were eligible for isolated CABG or PCI and 219 were included in the MASS V trial. All patients underwent resting cardiac magnetic resonance imaging. Of a total of 840 myocardial segments, 654 were nonischemic segments and 186 were ischemic segments. Native T1 and ECV values of ischemic segments were not significantly different from nonischemic segments, both at rest and after stress induction. In addition, native T1 and ECV values of myocardial segments supplied by vessels with obstructive lesions were similar to those supplied by nonobstructive ones. CONCLUSION AND RELEVANCE: In this study, cardiac magnetic resonance identified similar T1 mapping values between ischemic and nonischemic myocardial segments. This finding suggests integrity and stability of myocardial tissue in the presence of stress-induced ischemia.

Ischemic stroke caused by large-artery atherosclerosis: a red flag for subclinical coronary artery disease.

Background: The coronary calcium score (CAC) measured on chest computerized tomography is a risk marker of cardiac events and mortality. We compared CAC scores in two multiethnic groups without symptomatic coronary artery disease: subjects in the chronic phase after stroke or transient ischemic attack and at least one symptomatic stenosis ≥50% in the carotid or vertebrobasilar territories (Groupathero) and a control group (Groupcontrol). Methods: In this cross-sectional study, Groupathero included two subgroups: GroupExtraorIntra, with stenoses in either cervical or intracranial arteries, and GroupExtra&Intra, with stenoses in at least one cervical and one intracranial artery. Groupcontrol had no history of prior stroke/transient ischemic attacks and no stenoses ≥50% in cervical or intracranial arteries. Age and sex were comparable in all groups. Frequencies of CAC ≥100 and CAC > 0 were compared between Groupathero and Groupcontrol, as well as between GroupExtraorIntr, GroupExtra&Intra, and Groupcontrol, with bivariate logistic regressions. Multivariate analyses were also performed. Results: A total of 120 patients were included: 80 in Groupathero and 40 in Groupcontrol. CAC >0 was significantly more frequent in Groupathero (85%) than Groupcontrol (OR, 4.19; 1.74-10.07; p = 0.001). Rates of CAC ≥100 were not significantly different between Groupathero and Groupcontrol but were significantly greater in GroupExtra&Intra (n = 13) when compared to Groupcontrol (OR 4.67; 1.21-18.04; p = 0.025). In multivariate-adjusted analyses, "Groupathero" and "GroupExtra&Intra" were significantly associated with CAC. Conclusion: The frequency of coronary calcification was higher in subjects with stroke caused by large-artery atherosclerosis than in controls.

The Role of Radiomics in Rectal Cancer.

PURPOSE: Radiomics is a promising method for advancing imaging assessment in rectal cancer. This review aims to describe the emerging role of radiomics in the imaging assessment of rectal cancer, including various applications of radiomics based on CT, MRI, or PET/CT. METHODS: We conducted a literature review to highlight the progress of radiomic research to date and the challenges that need to be addressed before radiomics can be implemented clinically. RESULTS: The results suggest that radiomics has the potential to provide valuable information for clinical decision-making in rectal cancer. However, there are still challenges in terms of standardization of imaging protocols, feature extraction, and validation of radiomic models. Despite these challenges, radiomics holds great promise for personalized medicine in rectal cancer, with the potential to improve diagnosis, prognosis, and treatment planning. Further research is needed to validate the clinical utility of radiomics and to establish its role in routine clinical practice. CONCLUSION: Overall, radiomics has emerged as a powerful tool for improving the imaging assessment of rectal cancer, and its potential benefits should not be underestimated.

Chronic troponin elevation assessed by myocardial T1 mapping in patients with stable coronary artery disease.

BACKGROUND: Cardiac troponin detected with sensitive assays can be chronically elevated, in the absence of unstable coronary syndromes. In patients with chronic coronary artery disease, clinically silent ischemic episodes may cause chronic troponin release. T1 mapping is a cardiovascular magnetic resonance technique useful in quantitative cardiac tissue characterization. We selected patients with anatomically and functionally normal hearts to investigate associations between chronic troponin release and myocardial tissue characteristics assessed by T1 mapping. METHODS: We investigated the relationship between cardiac troponin I concentrations and cardiovascular magnetic resonance T1 mapping parameters in patients with stable coronary artery disease enrolled in MASS V study before elective revascularization. Participants had no previous myocardial infarction, negative late gadolinium enhancement, normal left ventricular function, chamber dimensions and wall thickness. RESULTS: A total of 56 patients were analyzed in troponin tertiles: nativeT1 and extracellular volume (ECV) values (expressed as means ±â€…standard deviations) increased across tertiles: nativeT1 (1006 ±â€…27 ms vs 1016 ±â€…27 ms vs 1034 ±â€…37 ms, ptrend = 0.006) and ECV (22 ±â€…3% vs 23 ±â€…1.9% vs 25 ±â€…3%, ptrend = 0.007). Cardiac troponin I concentrations correlated with native T1(R = 0.33, P = .012) and ECV (R = 0.3, P = .025), and were independently associated with nativeT1 (P = .049) and ventricular mass index (P = .041) in multivariable analysis. CONCLUSION: In patients with chronic coronary artery disease and structurally normal hearts, troponin I concentrations correlated with T1 mapping parameters, suggesting that diffuse edema or fibrosis scattered in normal myocardium might be associated with chronic troponin release.

MRI-based radiomic score increased mrTRG accuracy in predicting rectal cancer response to neoadjuvant therapy.

PURPOSE: To develop a magnetic resonance imaging (MRI)-based radiomics score, i.e., "rad-score," and to investigate the performance of rad-score alone and combined with mrTRG in predicting pathologic complete response (pCR) in patients with locally advanced rectal cancer following neoadjuvant chemoradiation therapy. METHODS: This retrospective study included consecutive patients with LARC who underwent neoadjuvant chemoradiotherapy followed by surgery from between July 2011 to November 2015. Volumes of interest of the entire tumor on baseline rectal MRI and of the tumor bed on restaging rectal MRI were manually segmented on T2-weighted images. The radiologist also provided the ymrTRG score on the restaging MRI. Radiomic score (rad-score) was calculated and optimal cut-off points for both mrTRG and rad-score to predict pCR were selected using Youden's J statistic. RESULTS: Of 180 patients (mean age = 63 years; 60% men), 33/180 (18%) achieved pCR. High rad-score (> - 1.49) yielded an area under the curve (AUC) of 0.758, comparable to ymrTRG 1-2 which yielded an AUC of 0.759. The combination of high rad-score and ymrTRG 1-2 yielded a significantly higher AUC of 0.836 compared with ymrTRG 1-2 and high rad-score alone (p < 0.001). A logistic regression model incorporating both high rad-score and mrTRG 1-2 was built to calculate adjusted odds ratios for pCR, which was 4.85 (p < 0.001). CONCLUSION: Our study demonstrates that a rectal restaging MRI-based rad-score had comparable diagnostic performance to ymrTRG. Moreover, the combined rad-score and ymrTRG model yielded a significant better diagnostic performance for predicting pCR.

Current status and future perspectives of radiomics in hepatocellular carcinoma.

Given the frequent co-existence of an aggressive tumor and underlying chronic liver disease, the management of hepatocellular carcinoma (HCC) patients requires experienced multidisciplinary team discussion. Moreover, imaging plays a key role in the diagnosis, staging, restaging, and surveillance of HCC. Currently, imaging assessment of HCC entails the assessment of qualitative characteristics which are prone to inter-reader variability. Radiomics is an emerging field that extracts high-dimensional mineable quantitative features that cannot be assessed visually with the naked eye from medical imaging. The main potential applications of radiomic models in HCC are to predict histology, response to treatment, genetic signature, recurrence, and survival. Despite the encouraging results to date, there are challenges and limitations that need to be overcome before radiomics implementation in clinical practice. The purpose of this article is to review the main concepts and challenges pertaining to radiomics, and to review recent studies and potential applications of radiomics in HCC.

Long-term respiratory follow-up of ICU hospitalized COVID-19 patients: Prospective cohort study.

BACKGROUND: Coronavirus disease (COVID-19) survivors exhibit multisystemic alterations after hospitalization. Little is known about long-term imaging and pulmonary function of hospitalized patients intensive care unit (ICU) who survive COVID-19. We aimed to investigate long-term consequences of COVID-19 on the respiratory system of patients discharged from hospital ICU and identify risk factors associated with chest computed tomography (CT) lesion severity. METHODS: A prospective cohort study of COVID-19 patients admitted to a tertiary hospital ICU in Brazil (March-August/2020), and followed-up six-twelve months after hospital admission. Initial assessment included: modified Medical Research Council dyspnea scale, SpO2 evaluation, forced vital capacity, and chest X-Ray. Patients with alterations in at least one of these examinations were eligible for CT and pulmonary function tests (PFTs) approximately 16 months after hospital admission. Primary outcome: CT lesion severity (fibrotic-like or non-fibrotic-like). Baseline clinical variables were used to build a machine learning model (ML) to predict the severity of CT lesion. RESULTS: In total, 326 patients (72%) were eligible for CT and PFTs. COVID-19 CT lesions were identified in 81.8% of patients, and half of them showed mild restrictive lung impairment and impaired lung diffusion capacity. Patients with COVID-19 CT findings were stratified into two categories of lesion severity: non-fibrotic-like (50.8%-ground-glass opacities/reticulations) and fibrotic-like (49.2%-traction bronchiectasis/architectural distortion). No association between CT feature severity and altered lung diffusion or functional restrictive/obstructive patterns was found. The ML detected that male sex, ICU and invasive mechanic ventilation (IMV) period, tracheostomy and vasoactive drug need during hospitalization were predictors of CT lesion severity(sensitivity,0.78±0.02;specificity,0.79±0.01;F1-score,0.78±0.02;positive predictive rate,0.78±0.02; accuracy,0.78±0.02; and area under the curve,0.83±0.01). CONCLUSION: ICU hospitalization due to COVID-19 led to respiratory system alterations six-twelve months after hospital admission. Male sex and critical disease acute phase, characterized by a longer ICU and IMV period, and need for tracheostomy and vasoactive drugs, were risk factors for severe CT lesions six-twelve months after hospital admission.

Contrast phase recognition in liver computer tomography using deep learning.

Hepatocellular carcinoma (HCC) has become the 4th leading cause of cancer-related deaths, with high social, economical and health implications. Imaging techniques such as multiphase computed tomography (CT) have been successfully used for diagnosis of liver tumors such as HCC in a feasible and accurate way and its interpretation relies mainly on comparing the appearance of the lesions in the different contrast phases of the exam. Recently, some researchers have been dedicated to the development of tools based on machine learning (ML) algorithms, especially by deep learning techniques, to improve the diagnosis of liver lesions in imaging exams. However, the lack of standardization in the naming of the CT contrast phases in the DICOM metadata is a problem for real-life deployment of machine learning tools. Therefore, it is important to correctly identify the exam phase based only on the image and not on the exam metadata, which is unreliable. Motivated by this problem, we successfully created an annotation platform and implemented a convolutional neural network (CNN) to automatically identify the CT scan phases in the HCFMUSP database in the city of São Paulo, Brazil. We improved this algorithm with hyperparameter tuning and evaluated it with cross validation methods. Comparing its predictions with the radiologists annotation, it achieved an accuracy of 94.6%, 98% and 100% in the testing dataset for the slice, volume and exam evaluation, respectively.

Forging Connections in Latin America to Advance AI in Radiology.

The 1° Encontro Latino-Americano de IA em Saúde (1st Latin American Meeting on AI in Health) was held during the 2022 Jornada Paulista de Radiologia, the annual radiology meeting in the state of São Paulo. The event was created to foster discussion among Latin American countries about the complexity, challenges, and opportunities in developing and using artificial intelligence (AI) in those countries. Technological improvements in AI have created high expectations in health care. AI is recognized increasingly as a game changer in clinical radiology. To counter the fear that AI would "take over" radiology, the program included activities to educate radiologists. The development of AI in Latin America is in its early days, and although there are some pioneer cases, many regions still lack world-class technological infrastructure and resources. Legislation, regulation, and public policies in data privacy and protection, digital health, and AI are recent advances in many countries. The meeting program was developed with a broad scope, with expertise from different countries, backgrounds, and specialties, with the objective of encompassing all levels of complexity (from basic concepts to advanced techniques), perspectives (clinical, technical, ethical, and business), and specialties (both informatics and data science experts and the usual radiology clinical groups). It was an opportunity to connect with peers from other countries and share lessons learned about AI in health care in different countries and contexts. Keywords: Informatics, Use of AI in Education, Impact of AI on Education, Social Implications © RSNA, 2022.