Search

The utility of three-dimensional (3D) printed models for medical education in complex congenital heart disease (CHD) is sparse and limited. The purpose of this study was to evaluate the utility of 3D printed models for medical education in criss-cross hearts covering a wide range of participants with different levels of knowledge and experience, from medical students, clinical fellows up to senior medical personnel. Study participants were enrolled from four dedicated imaging workshops developed between 2016 and 2019. The study design was a non-randomized cross-over study to evaluate 127 participants' level of understanding of the criss-cross heart anatomy. This was evaluated using the scores obtained following teaching with conventional images (echocardiography and magnetic resonance imaging) versus a 3D printed model learning approach. A significant improvement in anatomical knowledge of criss-cross heart anatomy was observed when comparing conventional imaging test scores to 3D printed model tests [76.9% (61.5%–87.8%) vs. 84.6% (76.9%–96.2%), P < 0.001]. The increase in the questionnaire marks was statistically significant across all academic groups (consultants in pediatric cardiology, fellows in pediatric cardiology, and medical students). Ninety-four percent (120) and 95.2% (121) of the participants agreed or strongly agreed, respectively, that 3D models helped them to better understand the medical images. Participants scored their overall satisfaction with the 3D printed models as 9.1 out of 10 points. In complex CHD such as criss-cross hearts, 3D printed replicas improve the understanding of cardiovascular anatomy. They enhanced the teaching experience especially when approaching medical students., The Association for European Pediatric and Congenital Cardiology Junior Research Grant 2015 to support the 3D printing scientific project in the field of pediatric and congenital cardiology in Europe (to Dr. Maria Nieves Velasco Forte), Grant Number: AEPC-IWG-JRG2015

[Background] Improvements in outcomes for patients with congenital heart disease (CHD) have increased the need for diagnostic and interventional procedures. Cumulative radiation risk is a growing concern. MRI-guided interventions are a promising ionizing radiation-free, alternative approach., [Purpose] To assess the feasibility of MRI-guided catheterization in young patients with CHD using advanced visualization passive tracking techniques., [Study Type] Prospective., [Population] A total of 30 patients with CHD referred for MRI-guided catheterization and pulmonary vascular resistance analysis (median age/weight: 4 years / 15 kg)., [Field Strength/Sequence] 1.5T; partially saturated (pSAT) real-time single-shot balanced steady-state free-precession (bSSFP) sequence., [Assessment] Images were visualized by a single viewer on the scanner console (interactive mode) or using a commercially available advanced visualization platform (iSuite, Philips). Image quality for anatomy and catheter visualization was evaluated by three cardiologists with >5 years' experience in MRI-catheterization using a 1–5 scale (1, poor, 5, excellent). Catheter balloon signal-to-noise ratio (SNR), blood and myocardium SNR, catheter balloon/blood contrast-to-noise ratio (CNR), balloon/myocardium CNR, and blood/myocardium CNR were measured. Procedure findings, feasibility, and adverse events were recorded. A fraction of time in which the catheter was visible was compared between iSuite and the interactive mode., [Statistical Tests] T-test for numerical variables. Wilcoxon signed rank test for categorical variables., [Results] Nine patients had right heart catheterization, 11 had both left and right heart catheterization, and 10 had single ventricle circulation. Nine patients underwent solely MRI-guided catheterization. The mean score for anatomical visualization and contrast between balloon tip and soft tissue was 3.9 ± 0.9 and 4.5 ± 0.7, respectively. iSuite provided a significant improvement in the time during which the balloon was visible in relation to interactive imaging mode (66 ± 17% vs. 46 ± 14%, P, [Data Conclusion] MRI-guided catheterizations were carried out safely and is feasible in children and adults with CHD. The pSAT sequence offered robust and simultaneous high contrast visualization of the catheter and cardiac anatomy.

Aims Superior sinus venosus atrial septal defect (SVASD) is commonly associated with partial anomalous pulmonary venous drainage (PAPVD). We aimed to describe the first series of percutaneous SVASD and PAPVD correction using a two-step simulation for procedural planning. Methods and results Patients with SVASD and right PAPVD with a clinical indication for correction were selected. They underwent an ex vivo procedural simulation on a 3D-printed model followed by an in vivo simulation using balloon inflation in the targeted stent landing zone. The percutaneous procedure consisted in deploying a 10-zig custom-made covered stent in the SVC-RA junction. Five patients were referred for preprocedural evaluation and were deemed suitable for percutaneous correction. The procedure was successful in all patients with no residual interatrial shunt and successful redirection of the pulmonary venous drainage to the left atrium. At a median clinical follow-up of 8.1 months (2.6-19.8), no adverse events were noted, and all patients showed clinical improvement. During follow-up, transthoracic echocardiography and multidetector cardiac tomography in four patients or invasive angiography in one patient demonstrated a patent SVC stent, and no residual SVASD and unobstructed PV drainage in all patients. Conclusions In selected patients using a two-stage simulation strategy, percutaneous correction of SVASD with PAPVD is feasible and safe, and led to favourable short-term outcomes.

Sinus venosus atrial septal defects (SVASD) with partial anomalous pulmonary venous drainage (PAPVD) are conventionally treated surgically. Cardiac magnetic resonance (CMR) ([Figure 1][1], [Online Video 1][2]), patient-specific 3-dimensional (3D) printing ([Online Appendix][3]) and in vitro

Robotic-assisted kidney transplantation (RAKT) offers key benefits for patients that have been demonstrated in several studies. A barrier to the wider uptake of RAKT is surgical skill acquisition. This is exacerbated by the challenges of modern surgery with reduced surgical training time, patient safety concerns and financial pressures. Simulation is a well-established method of developing surgical skill in a safe and controlled environment away from the patient. We have developed a 3D printed simulation model for the key step of the kidney transplant operation which is the vascular anastomosis. The model is anatomically accurate, based on the CT scans of patients and it incorporates deceased donor vascular tissue. Crucially, it was developed to be used in the robotic operating theatre with the operating robot to enhance its fidelity. It is portable and relatively inexpensive when compared with other forms of simulation such as virtual reality or animal lab training. It thus has the potential of being more accessible as a training tool for the safe acquisition of RAKT specific skills. We demonstrate this model here.

ObjectiveMitral valve anatomy has a significant impact on potential surgical options for patients with hypoplastic or borderline left ventricle. Papillary muscle morphology is a major component regarding this aspect. The purpose of this study was to use cardiac magnetic resonance to describe the differences in papillary muscle anatomy between normal, borderline, and hypoplastic left ventricles.MethodsWe carried out a retrospective, observational cardiac magnetic resonance study of children (median age 5.36 years) with normal (n=30), borderline (n=22), or hypoplastic (n=13) left ventricles. Borderline and hypoplastic cases had undergone an initial hybrid procedure. Morphological features of the papillary muscles, location, and arrangement were analysed and compared across groups.ResultsAll normal ventricles had two papillary muscles with narrow pedicles; however, 18% of borderline and 46% of hypoplastic cases had a single papillary muscle, usually the inferomedial type. In addition, in borderline or hypoplastic ventricles, the supporting pedicle occasionally displayed a wide insertion along the ventricular wall. The length ratio of the superolateral support was significantly different between groups (normal: 0.46±0.08; borderline: 0.39±0.07; hypoplastic: 0.36±0.1; p=0.009). No significant difference, however, was found when analysing the inferomedial type (0.42±0.09; 0.38±0.07; 0.39±0.22, p=0.39). The angle subtended between supports was also similar among groups (113°±17°; 111°±51° and 114°±57°; p=0.99). A total of eight children with borderline left ventricle underwent biventricular repair. There were no significant differentiating features for papillary muscle morphology in this subgroup.ConclusionsThe superolateral support can be shorter or absent in borderline or hypoplastic left ventricle cases. The papillary muscle pedicles in these patients often show a broad insertion. These changes have important implications on surgical options and should be described routinely.

Background Ex-vivo cardiovascular magnetic resonance (CMR) imaging has played an important role in the validation of in-vivo CMR characterization of pathological processes. However, comparison between in-vivo and ex-vivo imaging remains challenging due to shape changes occurring between the two states, which may be non-uniform across the diseased heart. A novel two-step process to facilitate registration between ex-vivo and in-vivo CMR was developed and evaluated in a porcine model of chronic myocardial infarction (MI). Methods Seven weeks after ischemia-reperfusion MI, 12 swine underwent in-vivo CMR imaging with late gadolinium enhancement followed by ex-vivo CMR 1 week later. Five animals comprised the control group, in which ex-vivo imaging was undertaken without any support in the LV cavity, 7 animals comprised the experimental group, in which a two-step registration optimization process was undertaken. The first step involved a heart specific flexible 3D printed scaffold generated from in-vivo CMR, which was used to maintain left ventricular (LV) shape during ex-vivo imaging. In the second step, a non-rigid co-registration algorithm was applied to align in-vivo and ex-vivo data. Tissue dimension changes between in-vivo and ex-vivo imaging were compared between the experimental and control group. In the experimental group, tissue compartment volumes and thickness were compared between in-vivo and ex-vivo data before and after non-rigid registration. The effectiveness of the alignment was assessed quantitatively using the DICE similarity coefficient. Results LV cavity volume changed more in the control group (ratio of cavity volume between ex-vivo and in-vivo imaging in control and experimental group 0.14 vs 0.56, p p p = 0.034). Following the non-rigid co-registration step of the process, the DICE similarity coefficients for the myocardium, LV cavity and scar were 0.93 (±0.02), 0.89 (±0.01) and 0.77 (±0.07) respectively and the myocardial tissue and LV cavity volumes had a ratio of 1.03 and 1.00 respectively. Conclusions The pattern of the morphological changes seen between the in-vivo and the ex-vivo LV differs between scar and healthy myocardium. A 3D printed flexible scaffold based on the in-vivo shape of the LV cavity is an effective strategy to minimize morphological changes in the ex-vivo LV. The subsequent non-rigid registration step further improved the co-registration and local comparison between in-vivo and ex-vivo data.

Advances in biomedical engineering have led to three-dimensional (3D)-printed models being used for a broad range of different applications. Teaching medical personnel, communicating with patients and relatives, planning complex heart surgery, or designing new techniques for repair of CHD via cardiac catheterisation are now options available using patient-specific 3D-printed models. The management of CHD can be challenging owing to the wide spectrum of morphological conditions and the differences between patients. Direct visualisation and manipulation of the patients’ individual anatomy has opened new horizons in personalised treatment, providing the possibility of performing the whole procedure in vitro beforehand, thus anticipating complications and possible outcomes. In this review, we discuss the workflow to implement 3D printing in clinical practice, the imaging modalities used for anatomical segmentation, the applications of this emerging technique in patients with structural heart disease, and its limitations and future directions., MNV was awarded the Association for European Paediatric and Congenital Cardiology Junior research grant on 2015 to support the 3D printing scientific project in the field of paediatric and congenital cardiology in Europe.

The use of 3D printed models of the right ventricular outflow tract (RVOT) for surgical and interventional planning is growing and often requires image segmentation of cardiac magnetic resonance (CMR) images. Segmentation results may vary based on contrast, image sequence, signal threshold chosen by the operator, and manual post-processing. The purpose of this study was to determine potential biases and post-processing errors in image segmentation to enable informed decisions. Models of the RVOT and pulmonary arteries from twelve patients who had contrast enhanced CMR angiography with gadopentetate dimeglumine (GPD), gadofosveset trisodium (GFT), and a post-GFT inversion-recovery (IR) whole heart sequence were segmented, trimmed, and aligned by three operators. Geometric agreement and minimal RVOT diameters were compared between sequences and operators. To determine the contribution of threshold, interoperator variability was compared between models created by the same two operators using the same versus different thresholds. Geometric agreement by Dice between objects was high (intraoperator: 0.89-0.95; interoperator: 0.95-0.97), without differences between sequences. Minimal RVOT diameters differed on average by - 1.9 to - 1.3 mm (intraoperator) and by 0.4 to 1.4 mm (interoperator). The contribution of threshold to interoperator geometric agreement was not significant (same threshold: 0.96 ± 0.06, different threshold: 0.93 ± 0.05; p = 0.181), but minimal RVOT diameters were more variable with different versus constant thresholds (- 9.12% vs. 2.42%; p

Three-dimensional printing has an increasing number of clinical applications in pediatric cardiology. Time required for dataset segmentation and conversion to stereolithography (STL) format remains a significant limitation. We investigated the impact of semi-automated cardiovascular-specific segmentation software on time and reproducibility of segmentation. Magnetic resonance angiograms (MRAs) of 19 patients undergoing intervention for right ventricular outflow lesions were segmented to demonstrate the right heart. STLs were created by two independent clinicians using semi-automated cardiovascular segmentation (SAS) and traditional manual segmentation (MS). Time was recorded and geometric STL disagreement was determined (0 % = no disagreement, 100 % = complete disagreement). MRA datasets were categorized as clean when only right heart structures were present in the MRA, or contaminated when left heart structures were also present and required removal. Eighteen (seven clean and 11 contaminated) cases were successfully segmented with both methods. Time to STL for clean datasets was faster with MS than SAS [median 209 s (IQR 192–252) vs. 296 s (272–317), p = 0.018] while contaminated datasets were faster with SAS [455 s (384–561) vs. 866 s (310–1429), p = 0.033]. Interobserver STL geometric disagreement was significantly lower using SAS than MS overall (0.70 ± 1.15 % vs. 1.31 ± 1.52 %, p = 0.030), and for the contaminated subset (0.81 ± 1.08 % vs. 1.75 ± 1.57 %, p = 0.036). Most geometric disagreement occurred at areas where left heart contamination was removed. Semi-automated segmentation was faster and more reproducible for contaminated datasets, while MS was faster but equally reproducible for clean datasets. Semi-automated segmentation methods are preferable for contaminated datasets and continued refinement of these tools should be supported.

We present a novel method to explicitly incorporate topological prior knowledge into deep learning based segmentation, which is, to our knowledge, the first work to do so. Our method uses the concept of persistent homology, a tool from topological data analysis, to capture high-level topological characteristics of segmentation results in a way which is differentiable with respect to the pixelwise probability of being assigned to a given class. The topological prior knowledge consists of the sequence of desired Betti numbers of the segmentation. As a proof-of-concept we demonstrate our approach by applying it to the problem of left-ventricle segmentation of cardiac MR images of subjects from the UK Biobank dataset, where we show that it improves segmentation performance in terms of topological correctness without sacrificing pixelwise accuracy.

Objectives Stent implantation for the treatment of aortic coarctation has become a standard approach for the management of older children and adults. Criteria for optimal stent design and construction remain undefined. This study used computational modelling to compare the performance of two generations of the Cheatham-Platinum stent (NuMED, Hopkinton, NY, USA) deployed in aortic coarctation using finite element analysis. Design Three-dimensional models of both stents, reverse engineered from microCT scans, were implanted in the aortic model of one representative patient. They were virtually expanded in the vessel with a 16 mm balloon and a pressure of 2 atm. Results The conventional stent foreshortened to 96.5% of its initial length, whereas the new stent to 99.2% of its initial length. Diameters in 15 slices across the conventional stent were 11.6–15 mm (median 14.2 mm) and slightly higher across the new stent: 10.7–15.3 mm (median 14.5 mm) (p= 0.021). Apposition to the vessel wall was similar: conventional stent 31.1% and new stent 28.6% of total stent area. Conclusions The new design Cheatham-Platinum stent showed similar deployment results compared to the conventional design. The new stent design showed slightly higher expansion, using the same delivery balloon. Patient-specific computational models can be used for virtual implantation of new aortic stents and promise to inform subsequent in vivo trials.

Building on theory and research, this book aims to share good practices and innovative ideas which have been successfully applied in the area of videoconferencing in language teaching and learning in the context of higher education. It offers practical ideas and methodologies, advantages and challenges of new approaches, and the what, how and why of individual cases so that readers can apply them in their own contexts. It intends to help language teachers and teacher trainers develop skills and confidence and to inspire them to reflect on their own teaching practices. It also offers insights on the impact of videoconferencing practices on learners and learning processes.

Background Cardiac catheterization is a common procedure in patients with congenital heart disease (CHD). Although cardiovascular magnetic resonance imaging (CMR) represents a promising alternative approach to fluoroscopy guidance, simultaneous high contrast visualization of catheter, soft tissue and the blood pool remains challenging. In this study, a novel passive tracking technique is proposed for enhanced positive contrast visualization of gadolinium-filled balloon catheters using partial saturation (pSAT) magnetization preparation. Methods The proposed pSAT sequence uses a single shot acquisition with balanced steady-state free precession (bSSFP) readout preceded by a partial saturation pre-pulse. This technique was initially evaluated in five healthy subjects. The pSAT sequence was compared to conventional bSSFP images acquired with (SAT) and without (Non-SAT) saturation pre-pulse. Signal-to-noise ratio (SNR) of the catheter balloon, blood and myocardium and the corresponding contrast-to-noise ratio (CNR) are reported. Subjective assessment of image suitability for CMR-guidance and ideal pSAT angle was performed by three cardiologists. The feasibility of the pSAT sequence is demonstrated in two adult patients undergoing CMR-guided cardiac catheterization. Results The proposed pSAT approach provided better catheter balloon/blood contrast and catheter balloon/myocardium contrast than conventional Non-SAT sequences. It also resulted in better blood and myocardium SNR than SAT sequences. When averaged over all volunteers, images acquired with a pSAT angle of 20° to 40° enabled simultaneous visualization of the catheter balloon and the cardiovascular anatomy (blood and myocardium) and were found suitable for CMR-guidance in >93% of cases. The pSAT sequence was successfully used in two patients undergoing CMR-guided diagnostic cardiac catheterization. Conclusions The proposed pSAT sequence offers real-time, simultaneous, enhanced contrast visualization of the catheter balloon, soft tissues and blood. This technique provides improved passive tracking capabilities during CMR-guided catheterization in patients. Electronic supplementary material The online version of this article (doi:10.1186/s12968-017-0368-0) contains supplementary material, which is available to authorized users.

Aims Coronary artery fistulae represent one of the most challenging anatomical defects to define accurately. We aimed to investigate the additional benefit conferred by volume rendering of tomographic images and 3D printing for diagnosis and interventional planning. Methods and results Four cases of coronary fistulae were considered for transcatheter closure. Multidetector computed tomography (three cases) or cardiac magnetic resonance (one case) images were acquired and segmented using Mimics software. Each case was reviewed after incremental consideration of diagnostic resources: two cardiologists reported source and volume-rendered images; device closure was discussed by the interventional cardiology team. All diagnoses and planned management were reviewed after inspection of a 3D model. Using source images alone, both cardiologists correctly described the course and drainage in two out of four cases. Aided by volume rendering, this improved to three out of four cases. Inspection of the 3D printed model prompted the planned interventional approach and device sizing to be altered in two out of four cases. In one out of four cases, the intervention was abandoned after inspection of the 3D printed model. Conclusions Diagnosis and management of patients with coronary artery fistulae rely on detailed image analyses. 3D models add value when determining the feasibility of, and the approach to intervention in these cases.

OBJECTIVES: To evaluate the impact of 3D printed models (3D models) on surgical planning in complex congenital heart disease (CHD). METHODS: A prospective case-crossover study involving 10 international centres and 40 patients with complex CHD (median age 3 years, range 1 month-34 years) was conducted. Magnetic resonance imaging and computed tomography were used to acquire and segment the 3D cardiovascular anatomy. Models were fabricated by fused deposition modelling of polyurethane filament, and dimensions were compared with medical images. Decisions after the evaluation of routine clinical images were compared with those after inspection of the 3D model and intraoperative findings. Subjective satisfaction questionnaire was provided. RESULTS: 3D models accurately replicate anatomy with a mean bias of -0.27 ± 0.73 mm. Ninety-six percent of the surgeons agree or strongly agree that 3D models provided better understanding of CHD morphology and improved surgical planning. 3D models changed the surgical decision in 19 of the 40 cases. Consideration of a 3D model refined the planned biventricular repair, achieving an improved surgical correction in 8 cases. In 4 cases initially considered for conservative management or univentricular palliation, inspection of the 3D model enabled successful biventricular repair. CONCLUSIONS: 3D models are accurate replicas of the cardiovascular anatomy and improve the understanding of complex CHD. 3D models did not change the surgical decision in most of the cases (21 of 40 cases, 52.5% cases). However, in 19 of the 40 selected complex cases, 3D model helped redefining the surgical approach.

Purpose Custom-made surface mold applicators often allow more flexibility when carrying out skin brachytherapy, particularly for small treatment areas with high surface obliquity. They can, however, be difficult to manufacture, particularly if there is a lack of experience in superficial high-dose-rate brachytherapy techniques or with limited resources. Methods and Materials We present a novel method of manufacturing superficial brachytherapy applicators utilizing three-dimensional (3D)–printing techniques. We describe the treatment planning process and the process of applicator manufacture. Results The treatment planning process, with the introduction of a pre-plan, allows for an “ideal” catheter arrangement within an applicator to be determined, exploiting varying catheter orientations, heights, and curvatures if required. The pre-plan arrangement is then 3D printed to the exact specifications of the pre-plan applicator design. This results in improved target volume coverage and improved sparing of organs at risk. Conclusions Using a pre-plan technique for ideal catheter placement followed by automated 3D-printed applicator manufacture has greatly improved the entire process of superficial high-dose-rate brachytherapy treatment. We are able to design and manufacture flexible, well-fitting, superior quality applicators resulting in a more efficient and improved patient pathway and patient experience.

Background Shortcomings in existing methods of image segmentation preclude the widespread adoption of patient-specific 3D printing as a routine decision-making tool in the care of those with congenital heart disease. We sought to determine the range of cardiovascular segmentation methods and how long each of these methods takes. Methods A systematic review of literature was undertaken. Medical imaging modality, segmentation methods, segmentation time, segmentation descriptive quality (SDQ) and segmentation software were recorded. Results Totally 136 studies met the inclusion criteria (1 clinical trial; 80 journal articles; 55 conference, technical and case reports). The most frequently used image segmentation methods were brightness thresholding, region growing and manual editing, as supported by the most popular piece of proprietary software: Mimics (Materialise NV, Leuven, Belgium, 1992–2015). The use of bespoke software developed by individual authors was not uncommon. SDQ indicated that reporting of image segmentation methods was generally poor with only one in three accounts providing sufficient detail for their procedure to be reproduced. Conclusions and implication of key findings Predominantly anecdotal and case reporting precluded rigorous assessment of risk of bias and strength of evidence. This review finds a reliance on manual and semi-automated segmentation methods which demand a high level of expertise and a significant time commitment on the part of the operator. In light of the findings, we have made recommendations regarding reporting of 3D printing studies. We anticipate that these findings will encourage the development of advanced image segmentation methods.

Background Transplantation is the treatment of choice for paediatric renal recipients. However, there are increased challenges in small ( Methods We describe our management in five paediatric renal recipients with vascular anomalies (median age 7 years [IQR 4·5–13·0], median weight 18 kg [IQR 14·5–29·0]). We assessed the utility of 3D printing as a planning tool in four children with complex abnormalities (one retrospective case, three prospective cases) for whom implantation was uncertain as judged by conventional imaging. Surgically relevant donor and recipient anatomy was segmented from MRI or CT data (Mimics Medical v18.0, Materialise, Leuven, Belgium). The segmentation geometry derived from the extracted anatomical data was then exported in STL file format and physically fabricated with multimaterial, polyjet 3D printing technology (Objet500 Connex1, Objet-Stratasys). We assessed the value of models using questionnaires and geometric validation studies. Findings Four (80%) of five children survived after one death from sepsis (with a functioning graft). At the latest median follow-up of 19 months (IQR 10·5–83·0) renal allograft survival was 100% (death censored) with a median estimated glomerular filtration rate of 55 mL/min per 1·73 m 2 (IQR 45–66). We have previously classified these vascular anomolies on the basis of aortic and IVC patency (I=aorta patent, II=infrarenal segment occluded, III=suprarenal segment occluded, IV=all aorta occluded) and similarly for IVC patency (A–D). By independent questionnaire, all prospective 3D printed models were considered useful for preoperative planning, and thereby facilitated transplantation. In our retrospective proof of concept, Bland–Altman analysis found that the mean difference in vascular diameter between the printed model and segmentation geometry was −0·1 mm (95% CI −0·7 to 0·5), which was insignificant when compared with the measurement uncertainty (±0·4 mm) and the limits of surgical precision. All models showed geometrical consistency with preprinting designs and intraoperative anatomical correlation within surgical acceptance for crucial decision making. Interpretation Vascular anomalies do not necessarily preclude transplantation, and a classification system could guide management. Our feasibility study of patient-specific 3D printing suggests that cases classified as sufficiently complex can benefit from this technology. Patient-specific models provide the surgical team with the full, 3D, accessible, haptic, and spatial appreciation of anatomy that is crucial in surgical decision making and planning. This technology can inform the selection of suitable anastamosis sites in the presence of anomalies and the best surgical approach for implantation of an adult-sized kidney into a small child. Funding None.

Background The dual inversion recovery (dual IR) pulse sequence has recently been shown to improve blood suppression and infarct delineation in late gadolinium enhancement (LGE) images of myocardial infarction. This resulted in significantly lower inter-observer variability in manual outlining of scar and higher expert confidence in scar detection and transmurality when compared with conventional inversion recovery (IR) images. Computer algorithms have been shown to improve the accuracy of scar segmentation within IR images. We sought to develop and optimise a set of computer algorithms to quantify scar in both IR and dual IR images. Methods

We introduce a method for training neural networks to perform image or volume segmentation in which prior knowledge about the topology of the segmented object can be explicitly provided and then incorporated into the training process. By using the differentiable properties of persistent homology, a concept used in topological data analysis, we can specify the desired topology of segmented objects in terms of their Betti numbers and then drive the proposed segmentations to contain the specified topological features. Importantly this process does not require any ground-truth labels, just prior knowledge of the topology of the structure being segmented. We demonstrate our approach in three experiments. Firstly we create a synthetic task in which handwritten MNIST digits are de-noised, and show that using this kind of topological prior knowledge in the training of the network significantly improves the quality of the de-noised digits. Secondly we perform an experiment in which the task is segmenting the myocardium of the left ventricle from cardiac magnetic resonance images. We show that the incorporation of the prior knowledge of the topology of this anatomy improves the resulting segmentations in terms of both the topological accuracy and the Dice coefficient. Thirdly, we extend the method to 3D volumes and demonstrate its performance on the task of segmenting the placenta from ultrasound data, again showing that incorporating topological priors improves performance on this challenging task. We find that embedding explicit prior knowledge in neural network segmentation tasks is most beneficial when the segmentation task is especially challenging and that it can be used in either a semi-supervised or post-processing context to extract a useful training gradient from images without pixelwise labels., Comment: IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020