Development of deep learning segmentation models for coronary X-ray angiography: Quality assessment by a new global segmentation score and comparison with human performance

Introduction and objectivesAlthough automatic artificial intelligence (AI) coronary angiography (CAG) segmentation is arguably the first step toward future clinical application, it is underexplored. We aimed to (1) develop AI models for CAG segmentation and (2) assess the results using similarity scores and a set of criteria defined by expert physicians.MethodsPatients undergoing CAG were randomly selected in a retrospective study at a single center. Per incidence, an ideal frame was segmented, forming a baseline human dataset (BH), used for training a baseline AI model (BAI). Enhanced human segmentation (EH) was created by combining the best of both. An enhanced AI model (EAI) was trained using the EH. Results were assessed by experts using 11 weighted criteria, combined into a Global Segmentation Score (GSS: 0–100 points). Generalized Dice Score (GDS) and Dice Similarity Coefficient (DSC) were also used for AI models assessment.Results1664 processed images were generated. GSS for BH, EH, BAI and EAI were 96.9+/-5.7; 98.9+/-3.1; 86.1+/-10.1 and 90+/-7.6, respectively (95% confidence interval, p<0.001 for both paired and global differences). The GDS for the BAI and EAI was 0.9234±0.0361 and 0.9348±0.0284, respectively. The DSC for the coronary tree was 0.8904±0.0464 and 0.9134±0.0410 for the BAI and EAI, respectively. The EAI outperformed the BAI in all coronary segmentation tasks, but performed less well in some catheter segmentation tasks.ConclusionsWe successfully developed AI models capable of CAG segmentation, with good performance as assessed by all scores.     

Influence of minimally invasive endodontic access cavities and bonding status of resin composites on the mechanical property of endodontically-treated teeth: A finite element study

ObjectiveTo study the mechanical behavior of endodontically-treated teeth with minimally invasive endodontic access cavities and resin composite restorations under different bonding conditions using finite element analysis (FEA).MethodsFour Class-II endodontic access cavities including the mesio-occlusal minimally-invasive (MO-MIE), mesio-occlusal conventional (MO-CONV), disto-occlusal minimally-invasive (DO-MIE), and disto-occlusal conventional (DO-CONV) cavities were prepared in 3D-printed maxillary first molars. Each tooth was subjected to root canal preparation and scanned using micro-CT to provide a 3D structural model which was virtually restored with resin composite. An intact 3D-printed molar was used as control. FEA was conducted under a 250-N vertical load. Three different interfacial bonding conditions between dentin/enamel and resin composite were considered, i.e. fully bonded, partially debonded, and fully debonded. The maximum principal stress of dentin and the normal tensile stress at the interfaces were recorded. The risk factor of failure for each component was then calculated.ResultsIn the fully-bonded tooth, the dentin-composite interface showed significantly higher stress and a higher risk factor than dentin, indicating that debonding at the dentin-composite interface would occur prior to dentin fracture. With the dentin-composite interface debonded, the enamel-composite interface exhibited higher stress and a higher risk factor than dentin, indicating that debonding at the enamel-composite interface would occur next, also prior to dentin fracture. With the resin composite fully debonded from the tooth, stress in dentin increased significantly. Irrespective of the bonding status, the CONV groups exhibited higher median stresses in dentin than the MIE groups.SignificanceWithin the limitation of this study, it was shown that debonding of the resin composite restoration increased the stress in dentin and hence the risk of dentin fracture in endodontically-restored teeth. Minimally-invasive access cavities could better safeguard the fracture resistance of interproximally-restored teeth compared to conventional ones.     

Safety data from the MenB-FHbp clinical development program in healthy individuals aged 10 years and older

BackgroundThe MenB-FHbp vaccine (Trumenba®) is licensed in various countries for the prevention of meningococcal serogroup B disease in individuals ≥ 10 years of age. The clinical development program included 11 completed trials where, in each trial, MenB-FHbp had an acceptable safety profile after a primary vaccination series was administered to individuals 10–65 years of age. However, the detection of potential rare events was limited because of individual clinical trial size. The current safety analysis evaluates pooled reactogenicity and other adverse events (AEs) reported in these trials to identify new safety signals not detectable in individual trials.MethodsEleven trials contributed safety data, of which 10 recorded local and systemic reactogenicity events; 8 of the trials were controlled, and reactogenicity data were pooled for 7 of these 8 trials. Additional AE evaluations included immediate AEs (IAEs), medically attended AEs (MAEs), serious AEs (SAEs), newly diagnosed chronic medical conditions (NDCMCs), and autoimmune or neuroinflammatory conditions.ResultsLocal and systemic reactions were more frequent in the MenB-FHbp group (n = 15,294) compared with controls (n = 5509), although most reactions were transient and mild to moderate in severity. Frequencies of IAEs, SAEs, MAEs, NDCMCs, and autoimmune or neuroinflammatory conditions were similar between the MenB-FHbp and control groups.ConclusionsMenB-FHbp demonstrated a favorable safety and tolerability profile in the clinical development program of > 15,000 vaccine recipients ≥ 10 years of age. No new safety signals were identified in the pooled analysis compared with data from the individual trials. Continued postmarketing safety surveillance is important for the identification of rare events.Clinicaltrials.gov: NCT01299480; NCT000808028; NCT00879814; NCT00780806; NCT01352845; NCT01352793; NCT01461993; NCT01323270; NCT01830855; NCT01461980; NCT01768117.     

Optimizing outcomes of colorectal cancer surgery with robotic platforms

Advanced robotic technology makes it easier to perform total mesorectal excision procedures in the narrow pelvis for rectal cancer while maintaining the advantages of minimally invasive surgery. Robotic surgery for rectal cancer leads to lower conversion rates and faster recovery of urogenital function than conventional laparoscopic surgery. However, longer operative time and high cost are major weaknesses of robotic surgery. To date, most other short-term surgical outcomes, pathologic outcomes, and long-term oncologic outcomes of robotic surgery have not shown significant advantages over laparoscopic surgery. However, robotic surgery is still a valid and highly anticipated surgical approach for rectal cancer because it greatly reduces the surgeon's workload and learning curve. There are also advantages when robotic techniques are applied to technically demanding procedures such as lateral pelvic lymph node dissection or intersphincteric resection. The introduction of new surgical robot systems, including the da Vinci® SP system, is expected to expand the applications of robotic surgery and provide new advantages.     

Fungal infections following liver transplantation

With increasing morbidity and mortality from chronic liver disease and acute liver failure, the need for liver transplantation is on the rise. Most of these patients are extremely vulnerable to infections as they are immune-compromised and have other chronic co-morbid conditions. Despite the recent advances in practice and improvement in diagnostic surveillance and treatment modalities, a major portion of these patients continue to be affected by post-transplant infections. Of these, fungal infections are particularly notorious given their vague and insidious onset and are very challenging to diagnose. This mini-review aims to discuss the incidence of fungal infections following liver transplantation, the different fungi involved, the risk factors, which predispose these patients to such infections, associated diagnostic challenges, and the role of prophylaxis. The population at risk is increasingly old and frail, suffering from various other co-morbid conditions, and needs special attention. To improve care and to decrease the burden of such infections, we need to identify the at-risk population with more robust clinical and diagnostic parameters. A more robust global consensus and stringent guidelines are needed to fight against resistant microbes and maintain the longevity of current antimicrobial therapies.