Search

Background: The literature on artificial intelligence (AI) in surgery has advanced rapidly during the past few years. However, the published studies on AI are mostly reported by computer scientists using their own jargon which is unfamiliar to surgeons., Methods: A literature search was conducted in using PubMed following the preferred reporting items for systematic reviews and meta-analyses (PRISMA) statement. The primary outcome of this review is to provide a glossary with definitions of the commonly used AI terms in surgery to improve their understanding by surgeons., Results: One hundred ninety-five studies were included in this review, and 38 AI terms related to surgery were retrieved. Convolutional neural networks were the most frequently culled term by the search, accounting for 74 studies on AI in surgery, followed by classification task (n = 62), artificial neural networks (n = 53), and regression (n = 49). Then, the most frequent expressions were supervised learning (reported in 24 articles), support vector machine (SVM) in 21, and logistic regression in 16. The rest of the 38 terms was seldom mentioned., Conclusions: The proposed glossary can be used by several stakeholders. First and foremost, by residents and attending consultant surgeons, both having to understand the fundamentals of AI when reading such articles. Secondly, junior researchers at the start of their career in Surgical Data Science and thirdly experts working in the regulatory sections of companies involved in the AI Business Software as a Medical Device (SaMD) preparing documents for submission to the Food and Drug Administration (FDA) or other agencies for approval., (© 2022. The Author(s).)

Background: More than 60% of patients affected by pancreatic cancer are ≥ 65 years of age. Surgery represents the only potentially curative treatment for malignant pancreatic neoplasia and a useful treatment for benign diseases., Aim: To evaluate outcomes in elderly patients with ASA risk score 4 who underwent pancreatic resection compared to younger patients and elderly patients with lower anesthesiological risk., Methods: A consecutive series of 345 patients underwent pancreatic resection between 2010 and 2017 was reviewed. We compared three groups based on age at the time of surgery: < 65 years (group A), 65-74 years (group B), and ≥ 75 years (group C). Patients in group C were split into two subgroups, ASA 1-3 versus ASA 4, and compared., Results: Group A consisted of 117 (34%) patients, group B 128 (37%) patients, and group C 100 (29%) patients. Group C had a significantly higher incidence of comorbidity and ASA 4 status (p < 0.05), and of overall post-operative complications (p < 0.01), because of the higher incidence of post-operative medical complications. No differences in terms of overall surgical complications and post-operative mortality were reported. The mean overall survival was significantly lower for group C (p < 0.01), with no difference in mortality for cancer. Within group C, no differences were reported regarding surgical complications (p = 0.59), mortality (p = 0.34), and mean overall survival (p = 0.53) between ASA 1-3 and ASA 4 patients., Conclusions: Advanced age should not preclude elderly patients with pancreatic diseases from being treated surgically, and ASA 4 in subjects aged ≥ 75 years should not be an absolute contraindication.

Purpose: To describe a novel technique designed to safely and precisely deploy the Supera stent accurately at the ostium of the proximal superficial femoral artery (SFA) without compromising the profunda and common femoral arteries., Technique: After antegrade crossing of the chronic total occlusion (CTO) at the SFA ostium and accurate predilation of the entire SFA lesion, a retrograde arterial access is obtained. The Supera stent is navigated in retrograde fashion to position the first crown to be released just at the SFA ostium. Antegrade dilation is performed across the retrograde access site to obtain adequate hemostasis. The technique has been applied successfully in 21 patients (mean age 78.1±8.2 years; 13 men) with critical limb ischemia using retrograde Supera stenting from the proximal anterior tibial artery (n=6), the posterior tibial artery (n=2), retrograde stent puncture in the mid to distal SFA (n=2), the native distal SFA/proximal popliteal segment (n=6), and the distal anterior tibial artery (n=5). No complications were observed., Conclusion: Distal retrograde Supera stent passage and reverse deployment allow precise and safe Supera stenting at the SFA ostium.

Goal of this study was to asses the performance of Aperio computer-assisted analysis for HER2 immunohistochemical measurement in 292 equivocally (score2+) HercepTest immunoreactive breast cancer cases, evaluated by an experienced pathologist and analyzed with fluorescent in situ hybridization (FISH). The automatic Aperio categorization and the percentage of immunoreactive cells as evaluated by the computer and by the pathologist were recorded. Computer-assisted analysis classified 7 (2.4%) cases as negative (0), 136 (46.6%) as faintly positive (1+), 134 (40.5%) as moderately positive (2+), and 15 (5.1%) as strongly positive (3+). Correlative component analysis (CCA) classification is associated with Her2 amplification (P<0.0001). Compared with the human evaluation, automated CCA classification would save 157 (58%) FISH analyses, while not identifying 15 amplified cases (6% false-negative rate). The mean computer percentage value (CPV) is 18.44% standard deviation ±19.00 (range, 0.01 to 76.10). CPV and the pathologist percentage value are significantly associated and correlated (P<0.001) and have similar sensitivity and specificity in identifying Her2 FISH-amplified cases. CPV has a very low interobserver variation. The difference in CPV in amplified and nonamplified subgroups is statistically significant (P<0.001). Receiver operating characteristic analysis indicates that CPV is good at separating FISH nonamplified from amplified cases (P<0.001). The optimal cut-off value maximizing both sensitivity and specificity is 17.6% (sensitivity=73.3%, specificity=71.6%). Using a different cut-off value (2% of positive cells) we would have missed only 3 amplified cases (1% false-negative rate) while not submitting to FISH 52 cases (18% of the whole series). This false-negative rate is well below the expected false-negative rate usually observed in score 1 cases, supporting the use of CCA with a modified cut-off value in routine diagnostics for equivocally stained HER2 cases.