The Lifelong Impact of Artificial Intelligence and Clinical Prediction Models on Patients With Tetralogy of Fallot

Medical advancements in the diagnosis, surgical techniques, perioperative care, and continued care throughout childhood have transformed the outlook for individuals with tetralogy of Fallot (TOF), improving survival and shifting the perspective towards lifelong care. However, with a growing population of survivors, longstanding challenges have been accentuated, and new challenges have surfaced, necessitating a re-evaluation of TOF care. Availability of prenatal diagnostics, insufficient information from traditional imaging techniques, previously unforeseen medical complications, and debates surrounding optimal timing and indications for reintervention are among the emerging issues. To address these challenges, the integration of artificial intelligence and machine learning holds great promise as they have the potential to revolutionize patient management and positively impact lifelong outcomes for individuals with TOF. Innovative applications of artificial intelligence and machine learning have spanned across multiple domains of TOF care, including screening and diagnosis, automated image processing and interpretation, clinical risk stratification, and planning and performing cardiac interventions. By embracing these advancements and incorporating them into routine clinical practice, personalized medicine could be delivered, leading to the best possible outcomes for patients. In this review, we provide an overview of these evolving applications and emphasize the challenges, limitations, and future potential for integrating them into clinical care.

Video analysis of ex vivo beating hearts during preservation on the TransMedics® organ care system

Background

Reliable biomarkers for assessing the viability of the donor hearts undergoing ex vivo perfusion remain elusive. A unique feature of normothermic ex vivo perfusion on the TransMedics® Organ Care System (OCS™) is that the donor heart is maintained in a beating state throughout the preservation period. We applied a video algorithm for an in vivo assessment of cardiac kinematics, video kinematic evaluation (Vi.Ki.E.), to the donor hearts undergoing ex vivo perfusion on the OCS™ to assess the feasibility of applying this algorithm in this setting.

Methods

Healthy donor porcine hearts (n = 6) were procured from Yucatan pigs and underwent 2 h of normothermic ex vivo perfusion on the OCS™ device. During the preservation period, serial high-resolution videos were captured at 30 frames per second. Using Vi.Ki.E., we assessed the force, energy, contractility, and trajectory parameters of each heart.

Results

There were no significant changes in any of the measured parameters of the heart on the OCS™ device over time as judged by linear regression analysis. Importantly, there were no significant changes in contractility during the duration of the preservation period (time 0-30 min, 918 ± 430 px/s; time 31-60 min, 1,386 ± 603 px/s; time 61-90 min, 1,299 ± 617 px/s; time 91-120 min, 1,535 ± 728 px/s). Similarly, there were no significant changes in the force, energy, or trajectory parameters. Post-transplantation echocardiograms demonstrated robust contractility of each allograft.

Conclusion

Vi.Ki.E. assessment of the donor hearts undergoing ex vivo perfusion is feasible on the TransMedics OCS™, and we observed that the donor hearts maintain steady kinematic measurements throughout the duration.

Cell Shortening and Calcium Homeostasis Analysis in Adult Cardiomyocytes via a New Software Tool

Intracellular calcium (Ca2+) is the central regulator of heart contractility. Indeed, it couples the electrical signal, which pervades the myocardium, with cardiomyocytes contraction. Moreover, alterations in calcium management are the main factors contributing to the mechanical and electrical dysfunction observed in failing hearts. So, simultaneous analysis of the contractile function and intracellular Ca2+ is indispensable to evaluate cardiomyocytes activity. Intracellular Ca2+ variations and fraction shortening are commonly studied with fluorescent Ca2+ indicator dyes associated with microscopy techniques. However, tracking and dealing with multiple files manually is time-consuming and error-prone and often requires expensive apparatus and software. Here, we announce a new, user-friendly image processing and analysis tool, based on ImageJ-Fiji/MATLAB® software, to evaluate the major cardiomyocyte functional parameters. We succeeded in analyzing fractional cell shortening, Ca2+ transient amplitude, and the kinematics/dynamics parameters of mouse isolated adult cardiomyocytes. The proposed method can be applied to evaluate changes in the Ca2+ cycle and contractile behavior in genetically or pharmacologically induced disease models, in drug screening and other common applications to assess mammalian cardiomyocyte functions.

Artificial Intelligence Supports Decision Making during Open-Chest Surgery of Rare Congenital Heart Defects

The human right ventricle is barely monitored during open-chest surgery due to the absence of intraoperative imaging techniques capable of elaborating its complex function. Accordingly, artificial intelligence could not be adopted for this specific task. We recently proposed a video-based approach for the real-time evaluation of the epicardial kinematics to support medical decisions. Here, we employed two supervised machine learning algorithms based on our technique to predict the patients' outcomes before chest closure. Videos of the beating hearts were acquired before and after pulmonary valve replacement in twelve Tetralogy of Fallot patients and recordings were properly labeled as the "unhealthy" and "healthy" classes. We extracted frequency-domain-related features to train different supervised machine learning models and selected their best characteristics via 10-fold cross-validation and optimization processes. Decision surfaces were built to classify two additional patients having good and unfavorable clinical outcomes. The k-nearest neighbors and support vector machine showed the highest prediction accuracy; the patients' class was identified with a true positive rate ≥95% and the decision surfaces correctly classified the additional patients in the "healthy" (good outcome) or "unhealthy" (unfavorable outcome) classes. We demonstrated that classifiers employed with our video-based technique may aid cardiac surgeons in decision making before chest closure.

Right ventricular functional recovery depends on timing of pulmonary valve replacement in tetralogy of Fallot: a video kinematic study

OBJECTIVES

Indications for and timing of pulmonary valve replacement (PVR) after tetralogy of Fallot repair are controversial. Among magnetic resonance imaging indices proposed to time valve replacement, a right ventricular (RV) end-diastolic volume index greater than 160 ml/m2 is often used. Recent evidence suggests that this value may still identify patients with irreversible RV dysfunction, thus hindering recovery. Our goal was to define, using intraoperative video kinematic evaluation, whether a relationship exists between timing of PVR and early functional recovery after surgery.

METHODS

Between November 2016 and November 2018, a total of 12 consecutive patients aged 27.1 ± 19.1 years underwent PVR on average 22.2 ± 13.3 years after tetralogy of Fallot repair. Mean RV end-diastolic volume evident on the magnetic resonance images was 136.9 ± 35.7 ml/m2. Intraoperative cardiac kinematics were assessed by video kinematic evaluation via a high-speed camera acquiring videos at 200 fps before and after valve replacement.

RESULTS

Patients presenting with RV end-diastolic volume <147 ml/m2 were significantly younger (11.2 ± 5.0 vs 38.4 ± 17.0; P = 0.005) and had a shorter time interval to valve replacement (11.0 ± 5.2 vs 30.1 ± 11.3; P = 0.03). The entire population showed a moderate correlation among energy expenditure, cardiac fatigue, perimeter of contraction and preoperative RV end-diastolic volume index. Both groups showed a reduction in all kinematic parameters after PVR, but those with end-diastolic volume >147 ml/m2 showed an unpredictable outcome.

CONCLUSIONS

Video kinematic evaluation provides insight into intraoperative RV recovery in patients with tetralogy of Fallot undergoing PVR. Accordingly, functional recovery can be expected in patients with preoperative end-diastolic volume <147 ml/m2.