Machine learning applications in cardiac computed tomography: a composite systematic review.

Artificial intelligence and machine learning (ML) models are rapidly being applied to the analysis of cardiac computed tomography (CT). We sought to provide an overview of the contemporary advances brought about by the combination of ML and cardiac CT. Six searches were performed in Medline, Embase, and the Cochrane Library up to November 2021 for (i) CT-fractional flow reserve (CT-FFR), (ii) atrial fibrillation (AF), (iii) aortic stenosis, (iv) plaque characterization, (v) fat quantification, and (vi) coronary artery calcium score. We included 57 studies pertaining to the aforementioned topics. Non-invasive CT-FFR can accurately be estimated using ML algorithms and has the potential to reduce the requirement for invasive angiography. Coronary artery calcification and non-calcified coronary lesions can now be automatically and accurately calculated. Epicardial adipose tissue can also be automatically, accurately, and rapidly quantified. Effective ML algorithms have been developed to streamline and optimize the safety of aortic annular measurements to facilitate pre-transcatheter aortic valve replacement valve selection. Within electrophysiology, the left atrium (LA) can be segmented and resultant LA volumes have contributed to accurate predictions of post-ablation recurrence of AF. In this review, we discuss the latest studies and evolving techniques of ML and cardiac CT.

Single-lead ECGs (AliveCor) are a feasible, cost-effective and safer alternative to 12-lead ECGs in community diagnosis and monitoring of atrial fibrillation.

BACKGROUND: Community management of atrial fibrillation (AF) often requires the use of electrocardiographic (ECG) investigation. Patients discharged following treatment of AF with fast ventricular response (fast AF) can require numerous ECGs to monitor rate and/or rhythm control. Single-lead ECGs have been proposed as a more convenient and relatively accurate alternative to 12-lead ECGs for rate/rhythm management and also diagnosis of AF. We aimed to examine the feasibility of using the AliveCor single-lead ECG monitor for diagnosis and monitoring of AF in the community setting. METHODS: During the course of 6 months, this evaluation of a clinical service improvement pathway used the AliveCor in management of patients requiring (1) follow-up ECGs for AF with previously documented rapid ventricular rate or (2) ECG confirmation of rhythm where AF was suspected. Twelve AliveCor devices provided to the acute community medical team were used to produce 30 s ECG rhythm strips (iECG) that were electronically sent to an overreading physician. RESULTS: Seventy-four patients (mean age 82 years) were managed on this pathway. (1) The AliveCor was successfully used to monitor the follow-up of 37 patients with fast AF, acquiring a combined total of 113 iECGs (median 1.5 ±3.75 per patient). None of these patients required a subsequent 12-lead ECG and this approach saved an estimate of up to £134.49 per patient. (2) Of 53 patients with abnormal pulses, the system helped identify 8 cases of new onset AF and 19 cases of previously known AF that had reverted from sinus back into AF. CONCLUSIONS: We have demonstrated that the AliveCor system is a feasible, cost-effective, time-efficient and potentially safer alternative to serial 12-lead ECGs for community monitoring and diagnosis of AF.

Quantitative evaluation and comparison of coronary artery characteristics by 3D coronary volume reconstruction.

Non-atherosclerotic abnormalities of vessel calibre, aneurysm and ectasia, are challenging to quantify and are often overlooked in qualitative reporting. Utilising a novel 3-dimensional (3D) quantitative coronary angiography (QCA) application, we have evaluated the characteristics of normal, diabetic and aneurysmal or ectatic coronary arteries. We selected 131 individuals under 50 years-of-age, who had undergone coronary angiography for suspected myocardial ischaemia between 1st January 2011 and 31st December 2015, at the Bristol Heart Institute, Bristol, UK. This included 42 patients with angiographically normal coronary arteries, 36 diabetic patients with unobstructed coronaries, and 53 patients with abnormal coronary dilatation (aneurysm and ectasia). A total of 1105 coronary segments were analysed using QAngio XA 3D (Research Edition, Medis medical imaging systems, Leiden, The Netherlands). The combined volume of the major coronary arteries was significantly different between each group (1240 ± 476 mm3 diabetic group, 1646 ± 391 mm3 normal group, and 2072 ± 687 mm3 abnormal group). Moreover, the combined coronary artery volumes correlated with patient body surface area (r = 0.483, p < 0.01). Inter-observer variability was assessed and intraclass correlation coefficient of the total coronary artery volume demonstrated a low variability of 3D QCA (r = 0.996, p < 0.001). Dedicated 3D QCA facilitates reproducible coronary artery volume estimation and allows discrimination of normal and diseased vessels.

Implantable cardioverter defibrillators in the context of hypertrophic cardiomyopathy: a lesson in patient autonomy.

Hypertrophiccardiomyopathy (HCM) is common, whereas the decision not to have an implantable cardioverterdefibrillator (ICD) when probably falling into a 'high-risk' category is not. A solicitor aged 45 years attended the inherited cardiac conditions clinic for review of her HCM and discussion about ICD implantation for primary prevention of sudden cardiac death (SCD). Despite a predicted 7% risk of SCD within the next 5 years, according to the European Society of Cardiology endorsed HCM Risk-SCD risk stratification tool, the patient opted against implantation of an ICD and comprehensively justifies her decision. This report discusses ethical aspects of a consultation offering ICD protection against SCD in the context of HCM and emphasises the clinicians' role in respecting patient autonomy.

Congenital talipes equinovarus and congenital vertical talus secondary to sacral agenesis.

Sacral agenesis is a rare congenital defect which is associated with foot deformities such as congenital talipes equinovarus (CTEV) and less commonly congenital vertical talus (CVT). We report a 3-year-old Caucasian girl who was born with right CTEV and left CVT secondary to sacral agenesis. Her right foot was managed with a Ponseti casting method at 2 weeks, followed by an Achilles tenotomy at 4 months. The left foot was initially managed with a nocturnal dorsi-flexion splint. Both feet remained resistant and received open foot surgery at 10 months producing plantigrade feet with neutral hindfeet. At 19 months, she failed to achieve developmental milestones and examinations revealed abnormal lower limb reflexes. A full body MRI was performed which identified the sacral agenesis. We advocate early MRI of the spine to screen for spinal defects when presented with resistant foot deformities, especially when bilateral.