Artificial intelligence methods for improved detection of undiagnosed heart failure with preserved ejection fraction.

AIM: Heart failure with preserved ejection fraction (HFpEF) remains under-diagnosed in clinical practice despite accounting for nearly half of all heart failure (HF) cases. Accurate and timely diagnosis of HFpEF is crucial for proper patient management and treatment. In this study, we explored the potential of natural language processing (NLP) to improve the detection and diagnosis of HFpEF according to the European Society of Cardiology (ESC) diagnostic criteria. METHODS AND RESULTS: In a retrospective cohort study, we used an NLP pipeline applied to the electronic health record (EHR) to identify patients with a clinical diagnosis of HF between 2010 and 2022. We collected demographic, clinical, echocardiographic and outcome data from the EHR. Patients were categorized according to the left ventricular ejection fraction (LVEF). Those with LVEF ≥50% were further categorized based on whether they had a clinician-assigned diagnosis of HFpEF and if not, whether they met the ESC diagnostic criteria. Results were validated in a second, independent centre. We identified 8606 patients with HF. Of 3727 consecutive patients with HF and LVEF ≥50% on echocardiogram, only 8.3% had a clinician-assigned diagnosis of HFpEF, while 75.4% met ESC criteria but did not have a formal diagnosis of HFpEF. Patients with confirmed HFpEF were hospitalized more frequently; however the ESC criteria group had a higher 5-year mortality, despite being less comorbid and experiencing fewer acute cardiovascular events. CONCLUSIONS: This study demonstrates that patients with undiagnosed HFpEF are an at-risk group with high mortality. It is possible to use NLP methods to identify likely HFpEF patients from EHR data who would likely then benefit from expert clinical review and complement the use of diagnostic algorithms.

An artificial intelligence tool for automated analysis of large-scale unstructured clinical cine cardiac magnetic resonance databases.

Aims: Artificial intelligence (AI) techniques have been proposed for automating analysis of short-axis (SAX) cine cardiac magnetic resonance (CMR), but no CMR analysis tool exists to automatically analyse large (unstructured) clinical CMR datasets. We develop and validate a robust AI tool for start-to-end automatic quantification of cardiac function from SAX cine CMR in large clinical databases. Methods and results: Our pipeline for processing and analysing CMR databases includes automated steps to identify the correct data, robust image pre-processing, an AI algorithm for biventricular segmentation of SAX CMR and estimation of functional biomarkers, and automated post-analysis quality control to detect and correct errors. The segmentation algorithm was trained on 2793 CMR scans from two NHS hospitals and validated on additional cases from this dataset (n = 414) and five external datasets (n = 6888), including scans of patients with a range of diseases acquired at 12 different centres using CMR scanners from all major vendors. Median absolute errors in cardiac biomarkers were within the range of inter-observer variability: <8.4 mL (left ventricle volume), <9.2 mL (right ventricle volume), <13.3 g (left ventricular mass), and <5.9% (ejection fraction) across all datasets. Stratification of cases according to phenotypes of cardiac disease and scanner vendors showed good performance across all groups. Conclusion: We show that our proposed tool, which combines image pre-processing steps, a domain-generalizable AI algorithm trained on a large-scale multi-domain CMR dataset and quality control steps, allows robust analysis of (clinical or research) databases from multiple centres, vendors, and cardiac diseases. This enables translation of our tool for use in fully automated processing of large multi-centre databases.

The Role of AI in Characterizing the DCM Phenotype.

Dilated Cardiomyopathy is conventionally defined by left ventricular dilatation and dysfunction in the absence of coronary disease. Emerging evidence suggests many patients remain vulnerable to major adverse outcomes despite clear therapeutic success of modern evidence-based heart failure therapy. In this era of personalized medical care, the conventional assessment of left ventricular ejection fraction falls short in fully predicting evolution and risk of outcomes in this heterogenous group of heart muscle disease, as such, a more refined means of phenotyping this disease appears essential. Cardiac MRI (CMR) is well-placed in this respect, not only for its diagnostic utility, but the wealth of information captured in global and regional function assessment with the addition of unique tissue characterization across different disease states and patient cohorts. Advanced tools are needed to leverage these sensitive metrics and integrate with clinical, genetic and biochemical information for personalized, and more clinically useful characterization of the dilated cardiomyopathy phenotype. Recent advances in artificial intelligence offers the unique opportunity to impact clinical decision making through enhanced precision image-analysis tasks, multi-source extraction of relevant features and seamless integration to enhance understanding, improve diagnosis, and subsequently clinical outcomes. Focusing particularly on deep learning, a subfield of artificial intelligence, that has garnered significant interest in the imaging community, this paper reviews the main developments that could offer more robust disease characterization and risk stratification in the Dilated Cardiomyopathy phenotype. Given its promising utility in the non-invasive assessment of cardiac diseases, we firstly highlight the key applications in CMR, set to enable comprehensive quantitative measures of function beyond the standard of care assessment. Concurrently, we revisit the added value of tissue characterization techniques for risk stratification, showcasing the deep learning platforms that overcome limitations in current clinical workflows and discuss how they could be utilized to better differentiate at-risk subgroups of this phenotype. The final section of this paper is dedicated to the allied clinical applications to imaging, that incorporate artificial intelligence and have harnessed the comprehensive abundance of data from genetics and relevant clinical variables to facilitate better classification and enable enhanced risk prediction for relevant outcomes.

Gastric perforation by a foreign body presenting as a pancreatic pseudotumour.

INTRODUCTION: Foreign body ingestion rarely causes complications, though it can pose a significant diagnostic challenge. Perforation, particularly of more muscular viscera, can present insidiously with a wide range of differential diagnoses. PRESENTATION OF CASE: Here we present a case of 75 year-old woman presenting with chest and epigastric pain. Initial imaging suggested a pancreatic lesion. Despite appropriate treatment she deteriorated clinically, and following urgent laparotomy a duck bone fragment was found to have perforated the lesser curvature of the stomach and embedded within the liver causing subhepatic abscess formation and associated inflammation. DISCUSSION: There are a number of examples of insidious presentations of gastrointestinal perforation. However, we have found only one other case of a perforation presenting as a pancreatic pseudotumour, and ours is the first to have been successfully managed by removal of the foreign body and drainage of the abscess alone. CONCLUSION: A high level of suspicion is required to make the correct diagnosis in cases such as these where the symptoms are not clear-cut. Thorough review and discussion of imaging prior to surgical treatment is essential to prevent unnecessary intervention.

A complex neurological presentation of syphilis.

Syphilis is a contagious sexually transmitted infection notable for its complex array of systemic presentations. It is caused by the spirochaete Treponema pallidum and although once considered to be a largely historical condition in the UK, the recent rise in incidence makes syphilis increasingly relevant when considering unusual presentations in at-risk patients. The disease has three stages: primary, secondary and tertiary. The tertiary stage is associated with a plethora of neurological features ranging from psychosis to seizure caused by direct invasion of the spirochaete into the central nervous system. Here we describe the case of a 45-year-old man presenting with tonic clonic seizures on a background of balance and visual problems. Following normal examination and routine investigations further serology confirmed a diagnosis of neurosyphilis. The patient was started on appropriate treatment and made an excellent clinical recovery.

Outcomes following the Kawashima procedure for single-ventricle palliation in left atrial isomerism.

OBJECTIVES: Patients with left atrial isomerism and interrupted inferior vena cava palliated with a superior cavopulmonary connection or Kawashima procedure (KP) have a high incidence of developing pulmonary arteriovenous malformations. The necessity for hepatic vein redirection (HVR) and its timing remains a controversy. We aimed to assess the clinical outcome of patients with left atrial isomerism following a KP. The main end points were death, requirement for HVR and the impact of HVR on oxygen saturation. METHODS: Retrospective review of 21 patients with a diagnosis of left atrial isomerism, interruption of the inferior vena cava and single-ventricle physiology managed with a KP at a single centre between January 1990 and March 2010. RESULTS: Twenty-one patients had a KP, with 12 subsequently undergoing HVR. There was relatively a constant monthly decrement in the proportion of patients who were free from death or HVR up until 60 months following the KP, with a dramatic increase in the hazard after this time. The Cox proportional hazards regression model demonstrated a reduced early risk for HVR or death in patients who underwent pulmonary artery banding versus arterial shunt as the primary procedure (hazard ratio: 0.10; P = 0.01), and an increased risk with bilateral superior vena cavas (SVCs) (hazard ratio: 3.4; P = 0.04) and age at KP (hazard ratio: 1.02 per month increase in age at KP; P = 0.02). HVR mortality was relatively high with 3 of 12 patients dying in the early postoperative period with profound cyanosis. The timing of HVR after the KP did not influence the postoperative rate of increase in oxygen saturation. CONCLUSION: These findings confirm that the majority of patients who undergo a KP will require HVR. Patients who are older at the time of the KP or having an initial arterial shunt or bilateral SVCs are at higher risk of HVR or death. The relatively high mortality at HVR was characterized by severe postoperative cyanosis.

Effect of noble gases on oxygen and glucose deprived injury in human tubular kidney cells.

The noble gas xenon has been shown to be protective in preconditioning settings against renal ischemic injury. The aims of this study were to determine the protective effects of the other noble gases, helium, neon, argon, krypton and xenon, on human tubular kidney HK2 cells in vitro. Cultured human renal tubular cells (HK2) were exposed to noble gas preconditioning (75% noble gas; 20% O(2); 5% CO(2)) for three hours or mock preconditioning. Twenty-four hours after gas exposure, cell injury was provoked with oxygen-glucose deprived (OGD) culture medium for three hours. Cell viability was assessed 24 h post-OGD by a 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide assay. Other cohorts of cultured cells were incubated in the absence of OGD in 75% noble gas, 20% O(2) and 5% CO(2) and cellular signals phospho-Akt (p-Akt), hypoxia-inducible factor-1alpha (HIF-1alpha) and Bcl-2 were assessed by Western blotting. OGD caused a reduction in cell viability to 0.382 +/- 0.1 from 1.0 +/- 0.15 at control (P < 0.01). Neon, argon and krypton showed no protection from injury (0.404 +/- 0.03; 0.428 +/- 0.02; 0.452 +/- 0.02; P > 0.05). Helium by comparison significantly enhanced cell injury (0.191 +/- 0.05; P < 0.01). Xenon alone exerted a protective effect (0.678 +/- 0.07; P < 0.001). In the absence of OGD, helium was also detrimental (0.909 +/- 0.07; P < 0.01). Xenon caused an increased expression of p-Akt, HIF-1alpha and Bcl-2, while the other noble gases did not modify protein expression. These results suggest that unlike other noble gases, preconditioning with the anesthetic noble gas xenon may have a role in protection against renal ischemic injury.

Neuroprotection (and lack of neuroprotection) afforded by a series of noble gases in an in vitro model of neuronal injury.

Xenon-induced neuroprotection has been well studied both in vivo and in vitro. In this study, the neuroprotective properties of the other noble gases, namely, krypton, argon, neon and helium, were explored in an in vitro model of neuronal injury. Pure neuronal cultures, derived from foetal BALB/c mice cortices, were provoked into injury by oxygen and glucose deprivation (OGD). Cultures were exposed to either nitrogen hypoxia or noble gas hypoxia in balanced salt solution devoid of glucose for 90min. The cultures were allowed to recover in normal culture medium for a further 24h in nitrogen or noble gas. The effect of noble gases on cell reducing ability in the absence of OGD was also investigated. Cell reducing ability was quantified via an MTT assay and expressed as a ratio of the control. The OGD caused a reduction in cell reducing ability to 0.56+/-0.04 of the control in the absence of noble gas (p<0.001). Like xenon (0.92+/-0.10; p<0.001), neuroprotection was afforded by argon (0.71+/-0.05; p<0.01). Neon and krypton did not have a protective effect under our experimental conditions. Helium had a detrimental effect on the cells. In the absence of OGD, krypton reduced the reducing ability of uninjured cells to 0.84+/-0.09 (p<0.01), but argon showed an improvement in reducing ability to 1.15+/-0.11 (p<0.05). Our data suggest that the cheap and widely available noble gas argon may have potential as a neuroprotectant for the future.