Advancement of artificial intelligence systems for surveillance endoscopy of Barrett's esophagus.

Barrett's esophagus (BE) is a precursor disease for esophageal adenocarcinoma. Timely detection and treatment has significant influence on patient outcomes. Over the last years, several artificial intelligence (AI) systems have emerged to assist the endoscopist. The primary focus of research has been computer aided detection (CADe). Several groups have succeeded in developing competitive models for neoplasia detection. Additionally, computer aided diagnosis (CADx) models have been developed for subsequent lesion characterization and assistance in clinical decision making. Future studies should focus on bridging the domain gap between academic development and integration in daily practice.

A deep learning system for detection of early Barrett's neoplasia: a model development and validation study.

BACKGROUND: Computer-aided detection (CADe) systems could assist endoscopists in detecting early neoplasia in Barrett's oesophagus, which could be difficult to detect in endoscopic images. The aim of this study was to develop, test, and benchmark a CADe system for early neoplasia in Barrett's oesophagus. METHODS: The CADe system was first pretrained with ImageNet followed by domain-specific pretraining with GastroNet. We trained the CADe system on a dataset of 14 046 images (2506 patients) of confirmed Barrett's oesophagus neoplasia and non-dysplastic Barrett's oesophagus from 15 centres. Neoplasia was delineated by 14 Barrett's oesophagus experts for all datasets. We tested the performance of the CADe system on two independent test sets. The all-comers test set comprised 327 (73 patients) non-dysplastic Barrett's oesophagus images, 82 (46 patients) neoplastic images, 180 (66 of the same patients) non-dysplastic Barrett's oesophagus videos, and 71 (45 of the same patients) neoplastic videos. The benchmarking test set comprised 100 (50 patients) neoplastic images, 300 (125 patients) non-dysplastic images, 47 (47 of the same patients) neoplastic videos, and 141 (82 of the same patients) non-dysplastic videos, and was enriched with subtle neoplasia cases. The benchmarking test set was evaluated by 112 endoscopists from six countries (first without CADe and, after 6 weeks, with CADe) and by 28 external international Barrett's oesophagus experts. The primary outcome was the sensitivity of Barrett's neoplasia detection by general endoscopists without CADe assistance versus with CADe assistance on the benchmarking test set. We compared sensitivity using a mixed-effects logistic regression model with conditional odds ratios (ORs; likelihood profile 95% CIs). FINDINGS: Sensitivity for neoplasia detection among endoscopists increased from 74% to 88% with CADe assistance (OR 2·04; 95% CI 1·73-2·42; p<0·0001 for images and from 67% to 79% [2·35; 1·90-2·94; p<0·0001] for video) without compromising specificity (from 89% to 90% [1·07; 0·96-1·19; p=0·20] for images and from 96% to 94% [0·94; 0·79-1·11; ] for video; p=0·46). In the all-comers test set, CADe detected neoplastic lesions in 95% (88-98) of images and 97% (90-99) of videos. In the benchmarking test set, the CADe system was superior to endoscopists in detecting neoplasia (90% vs 74% [OR 3·75; 95% CI 1·93-8·05; p=0·0002] for images and 91% vs 67% [11·68; 3·85-47·53; p<0·0001] for video) and non-inferior to Barrett's oesophagus experts (90% vs 87% [OR 1·74; 95% CI 0·83-3·65] for images and 91% vs 86% [2·94; 0·99-11·40] for video). INTERPRETATION: CADe outperformed endoscopists in detecting Barrett's oesophagus neoplasia and, when used as an assistive tool, it improved their detection rate. CADe detected virtually all neoplasia in a test set of consecutive cases. FUNDING: Olympus.

Multicenter study on the diagnostic performance of multiframe volumetric laser endomicroscopy targets for Barrett's esophagus neoplasia with histopathology correlation.

Volumetric laser endomicroscopy (VLE) has been shown to improve detection of early neoplasia in Barrett's esophagus (BE). However, diagnostic performance using histopathology-correlated VLE regions of interest (ROIs) has not been adequately studied. We evaluated the diagnostic accuracy of VLE assessors for identification of early BE neoplasia in histopathology-correlated VLE ROIs. In total, 191 ROIs (120 nondysplastic and 71 neoplastic) from 50 BE patients were evaluated in a random order using a web-based module. All ROIs contained histopathology correlations enabled by VLE laser marking. Assessors were blinded to endoscopic BE images and histology. ROIs were first scored as nondysplastic or neoplastic. Level of confidence was assigned to the predicted diagnosis. Outcome measures were: (i) diagnostic performance of VLE assessors for identification of BE neoplasia in all VLE ROIs, defined as accuracy, sensitivity, and specificity; (ii) diagnostic performance of VLE assessors for only high level of confidence predictions; and (iii) interobserver agreement. Accuracy, sensitivity, and specificity for BE neoplasia identification were 79% (confidence interval [CI], 75-83), 75% (CI, 71-79), and 81% (CI, 76-86), respectively. When neoplasia was identified with a high level of confidence, accuracy, sensitivity, and specificity were 88%, 83%, and 90%, respectively. The overall strength of interobserver agreement was fair (k = 0.29). VLE assessors can identify BE neoplasia with reasonable diagnostic accuracy in histopathology-correlated VLE ROIs, and accuracy is enhanced when BE neoplasia is identified with high level of confidence. Future work should focus on renewed VLE image reviewing criteria and real-time automatic assessment of VLE scans.

Improved Barrett's neoplasia detection using computer-assisted multiframe analysis of volumetric laser endomicroscopy.

Volumetric laser endomicroscopy (VLE) is a balloon-based technique, which provides a circumferential near-microscopic scan of the esophageal wall layers, and has potential to improve Barrett's neoplasia detection. Interpretation of VLE imagery in Barrett's esophagus (BE) however is time-consuming and complex, due to a large amount of visual information and numerous subtle gray-shaded VLE images. Computer-aided detection (CAD), analyzing multiple neighboring VLE frames, might improve BE neoplasia detection compared to automated single-frame analyses. This study is to evaluate feasibility of automatic data extraction followed by CAD using a multiframe approach for detection of BE neoplasia. Prospectively collected ex-vivo VLE images from 29 BE-patients with and without early neoplasia were retrospectively analyzed. Sixty histopathology-correlated regions of interest (30 nondysplastic vs. 30 neoplastic) were assessed using different CAD systems. Multiple neighboring VLE frames, corresponding to 1.25 millimeter proximal and distal to each region of interest, were evaluated. In total, 3060 VLE frames were analyzed via the CAD multiframe analysis. Multiframe analysis resulted in a significantly higher median AUC (median level = 0.91) compared to single-frame (median level = 0.83) with a median difference of 0.08 (95% CI, 0.06-0.10), P < 0.001. A maximum AUC of 0.94 was reached when including 22 frames on each side using a multiframe approach. In total, 3060 VLE frames were automatically extracted and analyzed by CAD in 3.9 seconds. Multiframe VLE image analysis shows improved BE neoplasia detection compared to single-frame analysis. CAD with multiframe analysis allows for fast and accurate VLE interpretation, thereby showing feasibility of automatic full scan assessment in a real-time setting during endoscopy.

Cost-effectiveness and cost-utility of endovascular versus open repair of ruptured abdominal aortic aneurysm in the Amsterdam Acute Aneurysm Trial.

BACKGROUND: Minimally invasive endovascular aneurysm repair (EVAR) could be a surgical technique that improves outcome of patients with ruptured abdominal aortic aneurysm (rAAA). The aim of this study was to analyse the cost-effectiveness and cost-utility of EVAR compared with standard open repair (OR) in the treatment of rAAA, with costs per 30-day and 6-month survivor as outcome parameters. METHODS: Resource use was determined from the Amsterdam Acute Aneurysm (AJAX) trial, a multicentre randomized trial comparing EVAR with OR in patients with rAAA. The analysis was performed from a provider perspective. All costs were calculated as if all patients had been treated in the same hospital (Onze Lieve Vrouwe Gasthuis, teaching hospital). RESULTS: A total of 116 patients were randomized. The 30-day mortality rate was 21 per cent after EVAR and 25 per cent for OR: absolute risk reduction (ARR) 4·4 (95 per cent confidence interval (c.i.) -11·0 to 19·7) per cent. At 6 months, the total mortality rate for EVAR was 28 per cent, compared with 31 per cent among those assigned to OR: ARR 2·4 (-14·2 to 19·0) per cent. The mean cost difference between EVAR and OR was €5306 (95 per cent c.i. -1854 to 12,659) at 30 days and €10,189 (-2477 to 24,506) at 6 months. The incremental cost-effectiveness ratio per prevented death was €120,591 at 30 days and €424,542 at 6 months. There was no significant difference in quality of life between EVAR and OR. Nor was EVAR superior regarding cost-utility. CONCLUSION: EVAR may be more effective for rAAA, but its increased costs mean that it is unaffordable based on current standards of societal willingness-to-pay for health gains.

Presence of (phospho)creatine in developing and adult skeletal muscle of mice without mitochondrial and cytosolic muscle creatine kinase isoforms.

We assessed the relationship between phosphocreatine (PCr) and creatine (Cr) content and creatine kinase (CK) activity in skeletal muscle of mice. The PCr and total Cr (tCr) concentrations, as well as CK activity, in hindlimb muscles of mice, with or without the cytosolic and mitochondrial isoforms of muscle creatine kinase (wild-type or CK--/-- mice), were determined by in vivo magnetic resonance (MR) spectroscopy and by biochemical means during postnatal growth and adulthood. In wild-type muscle the [tCr], PCr/ATP ratio and CK activity increased rapidly in the first 4-7 weeks. Remarkably, CK--/-- mice showed a similar increase in the PCr/ATP ratio during the first month in the presence of only minor brain-type BB-CK activity. Uptake of Cr in muscle was seemingly unrelated to CK activity as tCr increased in the same way in the muscles of both mouse types. At older ages the PCr/ATP ratio decreased in CK--/-- muscles, in contrast to wild-type where it still slowly increased, whereas [tCr] was similar for muscle of both mouse types. Using a new in vivo MR approach with application of [4-13C]Cr, a lower PCr/tCr ratio was also observed in CK--/-- muscle. From these data it follows that in vivo global ATP levels at rest are similar in the presence or absence of CK. Although Cr could still be converted to PCr in mature CK--/-- muscle, the immediate availability of PCr decreased, and PCr became partly inconvertible at older ages. Apparently, catalysis of the CK reaction by BB-CK, although significant in muscles of newborn mice, gradually declines to very low levels in adulthood. Part or all of this BB-CK may arise from satellite cells fusing with myotubes, a process that is most active during the first months of life. Finally, our observation that the MR and chemical assessment of muscle [tCr] and PCr/tCr ratio were similar for all mice does not support the existence of a significant MR-invisible or immobile pool of Cr, with a role for CK in this phenomenon.

Changes in mRNA expression profile underlie phenotypic adaptations in creatine kinase-deficient muscles.

We have studied the mechanisms that regulate the remodeling of the glycolytic, mitochondrial and structural network of muscles of creatine kinase M (M-CK)/sarcomeric mitochondrial creatine kinase (ScCKmit) knockout mice by comparison of wild-type and mutant mRNA profiles on cDNA arrays. The magnitudes of changes in mRNA levels were most prominent in M-CK/ScCKmit (CK(-/-)) double mutants but did never exceed those of previously observed changes in protein level for any protein examined. In gastrocnemius of CK(-/-) mice we measured a 2.5-fold increase in mRNA level for mitochondrial encoded cytochrome c oxidase (COX)-III which corresponds to the increase in protein content. The level of the nuclear encoded mRNAs for COX-IV, H(+)-ATP synthase-C, adenine nucleotide translocator-1 and insulin-regulatable glucose transporter-4 showed a 1.5-fold increase, also in agreement with protein data. In contrast, no concomitant up-regulation in mRNA and protein content was detected for the mitochondrial inorganic phosphate-carrier, voltage-dependent anion channel and certain glycolytic enzymes. Our results reveal that regulation of transcript level plays an important role, but it is not the only principle involved in the remodeling of mitochondrial and cytosolic design of CK(-/-) muscles.

Changes in glycolytic network and mitochondrial design in creatine kinase-deficient muscles.

Skeletal muscles respond with high plasticity to pathobiological conditions or changes in physiological demand by remodeling cytoarchitectural and metabolic characteristics of individual myocytes. We have previously shown that muscles of mice without mitochondrial and/or cytosolic creatine kinases (ScCKmit(-/-) and/or M-CK(-/-)) partly compensate for the defect(s) by redirecting metabolic pathways and ultrastructural characteristics. Here, we show by semiquantitative Western blot analysis that the compensatory changes involve mutation- and fiber-type-specific coordinated regulation of divergent but functionally coupled groups of proteins. Fast-twitch gastrocnemius muscle of CK(--/--) mice display a two- to fourfold upregulation of mitochondrial cytochrome c oxidase, inorganic phosphate carrier, adenine nucleotide translocator, and voltage-dependent anion channel proteins. In parallel, cytosolic myoglobin is upregulated. Slow-twitch soleus muscle responds with changes in the glycolytic enzyme pattern, including a shift in lactate dehydrogenase isoenzyme composition. Adaptations in the network for oxidative adenosine triphosphate (ATP) production are already apparent at 17 days of age.

The HMGI-C gene is a likely candidate for the autosomal dwarf locus in the chicken.

In order to map the autosomal dwarf (adw) locus in the chicken, 11 segregating families were created. Initially five of these families were used for a linkage experiment in which the genome was scanned with microsatellites using a technique called bulked segregant analysis. Subsequently animals from 11 families were typed individually for microsatellites that appeared to be linked. We were able to detect genetic linkage of the adw locus to five different microsatellite markers on chromosome 1, the closest showing a recombination fraction of only 0.03 (LOD score 32.12). In mice the phenotype pygmy shows a striking similarity to the autosomal dwarf phenotype in chickens, both having a disproportionately large head. The pygmy locus has been mapped on mouse chromosome 10 and found to represent a mutation in the gene coding for high-mobility group protein I-C (HMGI-C). Considering the synteny between regions of chicken chromosome 1, mouse chromosome 10 and human chromosome 12, and taking into account both the phenotypic characteristics and the mode of inheritance of the chicken adw and the mouse pygmy loci, the HMGI-C gene is a major candidate gene for the adw locus in the chicken. Fluorescence in situ hybridization of metaphase chromosomes with the chicken HMGI-C gene as a probe, showed that the chicken HMGI-C gene is indeed closely linked to marker LEI146 on chromosome 1.