Artificial Intelligence for Assessment of Endotracheal Tube Position on Chest Radiographs: Validation in Patients From Two Institutions.

BACKGROUND. Timely and accurate interpretation of chest radiographs obtained to evaluate endotracheal tube (ETT) position is important for facilitating prompt adjustment if needed. OBJECTIVE. The purpose of our study was to evaluate the performance of a deep learning (DL)-based artificial intelligence (AI) system for detecting ETT presence and position on chest radiographs in three patient samples from two different institutions. METHODS. This retrospective study included 539 chest radiographs obtained immediately after ETT insertion from January 1 to March 31, 2020, in 505 patients (293 men, 212 women; mean age, 63 years) from institution A (sample A); 637 chest radiographs obtained from January 1 to January 3, 2020, in 302 patients (157 men, 145 women; mean age, 66 years) in the ICU (with or without an ETT) from institution A (sample B); and 546 chest radiographs obtained from January 1 to January 20, 2020, in 83 patients (54 men, 29 women; mean age, 70 years) in the ICU (with or without an ETT) from institution B (sample C). A commercial DL-based AI system was used to identify ETT presence and measure ETT tip-to-carina distance (TCD). The reference standard for proper ETT position was TCD between greater than 3 cm and less than 7 cm, determined by human readers. Critical ETT position was separately defined as ETT tip below the carina or TCD of 1 cm or less. ROC analysis was performed. RESULTS. AI had sensitivity and specificity for identification of ETT presence of 100.0% and 98.7% (sample B) and 99.2% and 94.5% (sample C). AI had sensitivity and specificity for identification of improper ETT position of 72.5% and 92.0% (sample A), 78.9% and 100.0% (sample B), and 83.7% and 99.1% (sample C). At a threshold y-axis TCD of 2 cm or less, AI had sensitivity and specificity for critical ETT position of 100.0% and 96.7% (sample A), 100.0% and 100.0% (sample B), and 100.0% and 99.2% (sample C). CONCLUSION. AI identified improperly positioned ETTs on chest radiographs obtained after ETT insertion as well as on chest radiographs obtained of patients in the ICU at two institutions. CLINICAL IMPACT. Automated AI identification of improper ETT position on chest radiographs may allow earlier repositioning and thereby reduce complications.

Predictive Value of KLASS-02-QC Assessment Score on KLASS-02 Surgical Outcomes: Validation of Surgeon Quality Control and Standardization for D2 Lymphadenectomy.

OBJECTIVE: The aim of this study was to audit the 22 items and assessed each item's predictive value on surgical outcomes. BACKGROUND: The KLASS-02 trial revealed that the oncologic outcomes of laparoscopic distal gastrectomy are not inferior to open distal gastrectomy in patients with advanced gastric cancer. The surgeons participating in this trial were chosen based on the assessment scores from the KLASS-02-QC trial, which used 22 items for standardization of D2 lymphadenectomy and quality control. METHODS: We reviewed proficiency scores (PSs) for 22 items for 20 surgeons who participated in KLASS-02. The surgeons were divided into 2 groups according to PS, and the perioperative outcomes of 924 patients enrolled in KLASS-02 were compared between groups. Each item's predictive value for perioperative outcome was then assessed using multivariable regression models. RESULTS: Of the total 924 patients, 529 were operated on by high-score surgeons (high PS) and 395 were operated on by low-score surgeons (low-PS). High-PS group had less intraoperative blood loss, longer operation times, and fewer complications, major complications, reoperations, and shorter first flatus and hospital stay than low-PS group ( P =0.006, P <0.001, P <0.001, P <0.001, P =0.042, P =0.013, and P <0.001, respectively). Some items used in KLASS-02-QC predicted perioperative outcomes, such as intraoperative blood loss, major complications, reoperation, and hospital stay. CONCLUSIONS: Although this study only analyzed data associated with qualified surgeons, the 22 items effectively assessed the surgeons based on PS. A high score was associated with longer operation times, but better perioperative outcomes.

Assessment of laparoscopic stomach preserving surgery with sentinel basin dissection versus standard gastrectomy with lymphadenectomy in early gastric cancer-A multicenter randomized phase III clinical trial (SENORITA trial) protocol.

BACKGROUND: Along with the marked increase in early gastric cancer (EGC) in the Eastern countries, there has been an effort to adopt the sentinel node concept in EGC to preserve gastric function and reduce the occurrence of postoperative complications. Based on promising results from a previous quality control study, this prospective multicenter randomized controlled phase III clinical trial aims to elucidate the oncologic safety of laparoscopic stomach-preserving surgery with sentinel basin dissection (SBD) compared to a standard laparoscopic gastrectomy. METHODS/DESIGN: This trial is an investigator-initiated, open-label, multicenter randomized controlled phase III trial with a non-inferiority design. Patients diagnosed with a single lesion of clinical stage T1N0M0 gastric adenocarcinoma, with a diameter of 3 cm or less are eligible for the present study. A total of 580 patients (290 per group) will be randomized to either laparoscopic stomach-preserving surgery with SBD or standard surgery. The primary end-point is 3-year disease-free survival (DFS) and the secondary endpoints include postoperative morbidity and mortality, quality of life, 5-year DFS, and overall survival. Qualified investigators who completed the prior quality control study are exclusively allowed to participate in this phase III clinical trial. DISCUSSION: The proposed trial is expected to verify whether laparoscopic stomach-preserving surgery with SBD achieves similar oncologic outcomes and improved quality of life compared to a standard gastrectomy in EGC patients. TRIAL REGISTRATION: This study was registered at the NIH ClinicalTrial.gov database ( NCT01804998 ) on March 4th, 2013.

Multicenter Prospective Comparative Study of Robotic Versus Laparoscopic Gastrectomy for Gastric Adenocarcinoma.

OBJECTIVE: To compare short-term surgical outcomes including financial cost of robotic and laparoscopic gastrectomy. BACKGROUND: Despite a lack of supporting evidence, robotic surgery has been increasingly adopted as a minimally invasive modality for the treatment of gastric cancer because of its assumed technical superiority over conventional laparoscopy. METHODS: A prospective, multicenter comparative study was conducted. Patients were matched according to the surgeon, extent of gastric resection, and sex. The primary endpoint was morbidity and mortality. Outcomes were analyzed on an intention-to-treat and per-protocol basis. RESULTS: A total of 434 patients were enrolled for treatment with either robotic (n = 223) or laparoscopic (n = 211) gastrectomy for intention-to-treat analysis, and a total of 370 patients (n = 185 per treatment) were compared in per-protocol analysis. Results were similar between both analyses. In per-protocol analysis, both groups showed similar overall complication rates (robotic = 11.9% vs laparoscopic = 10.3%) and major complication rates (robotic = 1.1% vs laparoscopic = 1.1%) with no operative mortality in either group. Patients treated with robotic surgery showed significantly longer operative time (robotic = 221 minutes vs laparoscopic = 178 minutes; P < 0.001) and significantly higher total costs (robotic = US$13,432 vs laparoscopic = US$8090; P < 0.001), compared with those who underwent laparoscopic gastrectomy. No significant differences between groups were noted in estimated blood loss, rates of open conversion, diet build-up, or length of hospital stay. CONCLUSIONS: The use of robotic systems is assumed to provide a technically superior operative environment for minimally invasive surgery. However, our analysis of perioperative surgical outcomes indicated that robotic gastrectomy is not superior to laparoscopic gastrectomy. Clinical trials identification: NCT01309256.

The assessment of the oncological safety margin of insufficient lymph node dissection in pT2 (pm) gastric cancer.

PURPOSE: To identify the effect of insufficient lymph node dissection (LND) on the survival of patients with pT2 gastric cancer. MATERIALS AND METHODS: A total of 340 patients (120 patients with insufficient LND and others with D2 LND) who underwent gastrectomy for pT2 gastric cancer between January 2008 and December 2010 were included. RESULTS: The incidence of preoperatively diagnosed early gastric cancer was higher and there were fewer metastatic lymph nodes (LNs) in the insufficient LND group than the D2 group, but there was no survival difference between two groups (p=0.365). Among the 89 patients with metastatic LNs after D2 LND, 13 patients (14.6%) had metastatic LNs at selected N2 stations (#10, 11, or 12a), but none of these patients were in the pN1 category. One patient had five metastatic LNs at station #11p with no metastatic LNs at any other stations. The number of metastatic LNs was identified as the only risk factor for LN metastasis at selected N2 stations by logistic regression. CONCLUSION: If a patient has pN0 or pN1 category after insufficient LND for pT2 gastric cancer, the surgery can be regarded as secure. However, for patients with pN2 or pN3 category, more careful examination is required.

Assessment of open versus laparoscopy-assisted gastrectomy in lymph node-positive early gastric cancer: a retrospective cohort analysis.

BACKGROUND: Laparoscopy-assisted gastrectomy (LAG) is still limited for early gastric cancer (EGC) with low possibility of lymph node (LN) metastasis, due to the concern for incomplete LN dissection and controversial long-term outcomes. We assessed oncological outcomes of laparoscopy-assisted versus open gastrectomy (OG) for patients with LN positive EGC. METHODS: Between 2003 and 2007, 204 patients underwent surgery for LN positive EGC. We evaluated adequacy of LN dissection and early and long-term outcomes after OG (n = 162) and LAG (n = 42). RESULTS: Operative time was longer but hospital stay was shorter for LAG than OG. Postoperative complications occurred in 14 patients (8.6%) after OG and 1 patient (2.4%) after LAG (P = 0.316). Mean number of retrieved LNs and number of retrieved and metastatic LNs for each station did not differ between the two groups. During median 35 months of follow-up, 14 patients (8.6%) developed recurrence after OG, compared with 4 patients (9.5%) after LAG (P = 0.769). Overall 5-year disease-free survival was 89.9% and 89.7% after OG and LAG. Status of LN metastasis was the only independent prognostic factor for disease-free survival. CONCLUSIONS: LAG is an oncologically safe procedure even for LN positive EGC. Adequate LN dissection and comparable long-term outcomes to OG can be achieved by LAG.