Search

The existing deep learning based foreign object detection models for conveyor belts are relatively large and difficult to deploy on edge devices. There are errors and omissions in detecting foreign objects of different sizes and small objects. In order to solve the above problems, a foreign object detection method for coal mine conveyor belts based on improved YOLOv8 is proposed. The depthwise separable convolution, squeeze-and-excitation (SE) networks are used to reconstruct the Bottleneck of the C2f module in the YOLOv8 backbone network as a DSBlock, which improves the detection performance while keeping the model lightweight. To enhance the capability to obtain information from objects of different sizes, an efficient channel attention (ECA) mechanism is introduced. The input layer of ECA is subjected to adaptive average pooling and adaptive maximum pooling operations to obtain a cross channel interactive MECA module, which enhances the global visual information of the module and further improves the precision of foreign object recognition. The method modifies the 3 detection heads of YOLOv8 to 4 lightweight small object detection heads to enhance sensitivity to small objects and effectively reduce the missed and false detection rates of small object foreign objects. The experimental results show that the improved YOLOv8 achieves a precision of 91.69%, mAP@50 reached 92.27%, an increase of 3.09% and 4.07% respectively compared to YOLOv8. The detection speed of improved YOLOv8 reaches 73.92 frames/s, which can fully meet the demand for real-time detection of foreign objects on conveyor belts in coal mines. The improved YOLOv8 outperforms mainstream object detection algorithms such as SSD, Faster-RCNN, YOLOv5, and YOLOv7-tiny in terms of precision, mAP@50, number of parameters, weight size, and number of floating point operations.

Objective: To investigate the surgical efficacy and prognostic factors of T3NxM0 middle-low rectal cancer without neoadjuvant therapy., Methods: Clinical data of patients with middle-low rectal cancer undergoing TME surgery with T3NxM0 confirmed by postoperative pathology at Colorectal Surgery Department of Changhai Hospital from January 2008 to December 2010 were analyzed retrospectively., Inclusion Criteria: (1)no preoperative neoadjuvant chemoradiotherapy (nCRT); (2) complete preoperative evaluation, including medical history, preoperative colonoscopy or digital examination, blood tumor marker examination, and imaging examination; (3) distance between tumor lower margin and anal verge was ≤ 10 cm; (4) negative circumferential resection margin (CRM-). Finally, a total of 331 patients were included in this study. According to the number of metastatic lymph node confirmed by postoperative pathology, the patients were divided into N0 group without regional lymph node metastasis (190 cases) and N+ group with regional lymph node metastasis (141 cases). The perioperative conditions, local recurrence, distant metastasis and prognostic factors were analyzed., Results: Compared to N0 group in the perioperative data, N+ group had higher ratio of tumor deposit [29.8%(42/141) vs. 0, χ²=64.821, P<0.001] and vascular invasion [7.1%(10/141) vs. 0.5%(1/190),χ²=10.860, P<0.001]. There were no significant differences in tumor diameter, number of lymph nodes detected, positive nerve invasion, degree of tumor differentiation, morbidity of postoperative complication and postoperative adjuvant chemotherapy rate between the two groups (all P>0.05). The median follow-up period was 73.4 months. The merged 5-year local recurrence rate was 2.7%(9/331), 5-year distant metastasis rate was 23.3% (77/331), 5-year disease-free survival (DFS) rate was 73.4%, and 5-year overall survival (OS) rate was 77.2%. Multivariate analysis showed that lymph node metastasis (HR=3.120, 95%CI: 1.918 to 5.075, P<0.001), nerve invasion (HR=0.345, 95%CI: 0.156 to 0.760, P=0.008) and vascular invasion (HR=0.428, 95%CI: 0.189 to 0.972, P=0.043) were independent risk factors for DFS in patients with T3NxM0 rectal cancer after operation. Preoperative carcinoembryonic antigen level (HR=1.858, 95%CI:1.121 to 3.079, P=0.016), lymph node metastasis (HR=3.320, 95%CI: 1.985 to 5.553, P<0.001) and nerve invasion (HR=0.339, 95%CI: 0.156 to 0.738, P=0.006) were independent risk factors for OS in patients with T3NxM0 rectal cancer after operation., Conclusions: Optimal local control rate of middle-low rectal cancer patients with T3NxM0 and CRM- can be achieved by standard TME surgery alone. For patients with preoperative elevated blood carcinoembryonic antigen level, regional lymph node metastasis, or neurovascular invasion confirmed by pathology after surgery, adjuvant chemoradiotherapy should be actively applied after surgery to improve prognosis.

Objective: To summarize the diagnosis and treatment of iatrogenic colonoscopic perforation (ICP)., Methods: Clinical data, treatment course and outcome of 17 patients who developed ICP following colonoscopic examination or operation at Department of Colorectal Surgery, Changhai Hospital from January 2000 to December 2013 were retrospectively analyzed., Results: During above 13 years, a total of 127 106 patients underwent colonoscopic examination or operation, of whom 17 cases (0.013%) had ICP. There were 8 males and 9 females with an average age of 65.2 (32-85) years. The interval between the onset of ICP and clinically diagnosed ICP was 0 to 6 days after performance. ICP occurred in 8 patients following colonoscopy operations, including simple colonic polyp excision, endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD), while in 9 patients following simple colonoscopy examination. Except for one patient who was immediately diagnosed with ICP through the finding of "yellow adipose tissue visible in the vision field" during operation,7 early cases (41.2%) were diagnosed by abdominal X-ray examination, and 9 later cases were confirmed by abdominal CT examination. The perforation sites included sigmoid colon in 5 cases, caecum in 3 cases, descending colon in 3 cases, descending and sigmoid junction in 2 cases, ileum in 1 case, splenic flexure in 1 case, sigmoid and rectum junction in 1 case, retum in 1 case. One case with ICP following ESD after resection of polyp in caecum was cured successfully with conservative treatment, including fasting, gastrointestinal decompression, fluid infusion, anti-infection and nutritional support. One case with ICP, which was found during colonoscopic operation and the perforation was immediately closed with titanium clip, received conservative treatment, including anti-infection and then was cured. Fifteen patients underwent surgery, including 8 patients with primary intention intestinal perforation repair, 4 patients with primary intention resection of associated intestine and anastomosis, 2 patients with primary intention resection of associated intestine and ostomy, 1 patient with primary intention intestinal perforation repair and ostomy. Postoperative abdominal incision infection occurred in 4 cases, pulmonary infection in 1 case, incision infection with cardiovascular event or urinary tract infection in 1 case each. All the patients were cured and discharged. Average hospital stay was 18.6(3-45) days., Conclusions: ICP should be diagnosed by physical examination and imaging examination as soon as possible. For perforation during colonoscopic performance, colonoscopic titanium clip can be used for closure. Perforation repair is still the main procedure for ICP. If necessary, partial intestinal resection and anastomosis or ostomy can be selected.

Objective: To investigate the relationship between tumor regression grade (TRG) and lymph node regression grade (LRG) after neoadjuvant chemoradiotherapy (CRT) for rectal cancer and its clinical implication., Methods: Clinicopathological data of 176 rectal cancer patients undergoing radical excision after neoadjuvant CRT from January 2005 to December 2013 in our department were retrospectively analyzed., Inclusion Criteria: (1) Radiology indicated locally advanced low rectal cancer and patients had strong desire to preserve the sphincter before neoadjuvant CRT; (2) there was no definite metastatic lesion before neoadjuvant CRT; (3) patients received whole course of neoadjuvant CRT (regular radiotherapy plus synchronous fluorouracil-like drugs chemotherapy); (4) patients underwent radical operation after neoadjuvant CRT. Patients with short-course CRT and emergency surgery were excluded. TRG and LRG of postoperative specimens (including tumor and lymph nodes) were carried out based on the percentage of the fibrosis and the cancer residue. No cancer residue was defined as TRG1 and LRG1; rare cancer cell residue as TRG2 and LRG2; fibrosis growth over residual cancer as TRG3 and LRG3; residual cancer growth over fibrosis as TRG4 and LRG4; absence of regressive changes as TRG5 and LRG5; and normal lymph nodes as LRG0. Spearman correlation test was used to assess the correlation between TRG and LRG., Results: Of 176 patients, 111 were men and 65 were women. The mean age was (53.9±13.0) years. The number of patients with stage I(, II(, and III( before operation was 10, 49 and 62 while other 55 patients were unknown. Transabdominal low anterior resection (LAR) was performed in 118 cases and abdominal-perineal resection(APR) in 47 cases following the principle of total mesorectal excision (TME). Postoperative pathology of specimens revealed that the number of patients from TRG1 to TRG5 was 19 (10.8%), 25 (14.2%), 66 (37.5%), 47 (26.7%), 19 (10.8%), and from LRG0 to LRG5 was 35 (19.9%), 68 (38.6%), 10 (5.7%), 14 (8.0%), 15(8.5%), 34 (19.3%), respectively. TRG was correlated to LRG (P=0.005) while the Spearman correlation coefficient was only 0.24. The analysis of subgroup without LRG1 also showed that TRG was correlated to LRG(P=0.0005) and the Spearman correlation coefficient was 0.40., Conclusions: TRG can not represent LRG. Therefore, both TRG and LRG should be assessed when evaluating the response of rectal cancer to neoadjuvant CRT.

Objective: To compare the short-term outcomes between transrectal specimen extraction during laparoscopic sigmoid radical resection and conventional laparoscopy-assisted sigmoid radical resection., Methods: Sixteen patients(transrectal specimen extraction group,4 females and 12 males), who were planned to undergo laparoscopically assisted sigmoid radical resection with BMI<28 kg/m 2 and were evaluated as T1-T3 tumor by iconography without distant metastasis, were selected to undergo transrectal specimen extraction during laparoscopic sigmoid radical resection from December 2015 to April 2016 in the Department of Anorectal Surgery of Changhai Hospital. The procedure of specimen extraction was as follows: Perineal anal expansion was performed. The rectum was cut in rectal distal ligature within the abdominal cavity. Telescope cover was placed through Trocar hole in right low abdomen and rectal stump was pulled out of the body through the anus to form an access tunnel. Planned resected bowel was placed in the tunnel and the specimen was dissociated and removed completely from anus. Each patient in transrectal specimen extraction group was individually matched with two patients who underwent laparoscopically assisted sigmoid radical resection by gender, age, BMI and date of surgery. The perioperative outcomes and pathological evaluation of surgical specimen of two groups were retrospectively collected and compared., Results: The differences of baseline data (gender, age, BMI, distance from tumor to anal verge measured by colonoscopy and clinical tumor category) between two groups were not significant (all P>0.05). Compared to laparoscopy-assisted group, transrectal specimen extraction group presented longer operation time [(140.6±8.3) minutes vs. (122.2±26.2) minutes, t=-3.629, P=0.001], and more blood loss[(43.8±9.2) ml vs. (35.3±10.2) ml, t=-2.795, P=0.008], but shorter time to first flatus [(43.1±8.3) hours vs. (52.0±11.4) hours, t=2.756, P=0.008] and lower pain score at operative day and the first postoperative day (3.8±0.8 vs. 4.8±1.1, t=3.558, P=0.001; 2.6±0.6 vs. 3.8±0.8, t=5.165, P=0.000). The case ratio of additional analgesia [6.3%(1/16) ns. 18.8%(6/32)], postoperative hospital stay [(6.8±3.4) days vs. (5.6±0.8) days] and postoperative morbidity of complication [12.5%(2/16) vs. 9.4%(3/32)] were not significantly different between the two groups (all P>0.05). Within postoperative 30-day follow-up, transrectal specimen extraction group had ileus in one patient and anastomotic leakage in one patient, and laparoscopy-assisted group had fat necrosis of assisted incision in two patients and gastric retention in one patient. There were also no significant differences in specimen length[(18.2±4.8) cm vs. (19.8±5.7) cm, P>0.05], tumor size [(4.0±1.2) cm vs. (4.4±1.5) cm, P>0.05] and number of harvested lymph node (14.6±2.6 vs. 16.0±3.0, P>0.05] between two groups. During follow-up of 7-10(mean 9) months of transrectal specimen extraction group and 2-16 (mean 7) months of laparoscopically assisted group, no tumor local relapse and distant metastasis were found in the both groups., Conclusion: As compared to laparoscopy-assisted sigmoid radical resection, transrectal specimen extraction laparoscopic sigmoid radical resection has better short-term efficacy, meanwhile they have comparable oncologic clearance.

Objective: To identify the risk factors associated with lymph node metastasis in rectal cancer after neoadjuvant chemoradiotherapy (CRT)., Methods: From January 2005 to December 2013, the clinical data of 178 patients with advanced rectal cancer undergoing radical excision after neoadjuvant CRT in our department were reviewed retrospectively. A total of 11 clinicopathologic factors relating to lymph node metastasis were studied using univariate and multivariate Logistic regression analyses., Results: There were 74(41.6%) cases with lymph node metastasis, while 104 cases without lymph node metastasis. Univariate analysis showed that age(P=0.000 2), post-CRT CEA level(P=0.011 2), ypT stage(P=0.000 0), pathologic type(P=0.004 0), and tumor regression grade(TRG)(P=0.033 8) were significantly associated with lymph node metastasis. Multivariate analysis showed that age(OR=2.385, 95% CI:1.372 ~ 4.147, P=0.002 1), post-CRT CEA level(OR=2.310, 95% CI:1.005 ~ 5.307, P=0.048 6) and ypT stage(OR=2.592, 95% CI:1.236 ~ 5.432, P=0.011 7) were independent risk factors. However, 15.8% of the patients who achieved TRG1 had lymph node metastasis and TRG failed to independently correlate with lymph node metastasis in rectal cancer after neoadjuvant CRT., Conclusions: There was a higher ratio of lymph node metastasis in rectal cancer patients who were young, CEA≥5 μg/L or deep invasion after neoadjuvant CRT. Therefore, neoadjuvant CRT should be carefully considered in these patients.

Objective: To investigate the risk factors associated with lymph node metastasis of T1 and T2 rectal cancer., Methods: Clinicopathological data of 576 patients with stage T1 to T2 rectal cancer without serosal invasion confirmed by pathology undergoing curative resection in Changhai Hospital from January 1999 to December 2013 were analyzed retrospectively. The relationship of clinicopathological factors of overall patients and stage T1 patients with lymph node metastasis was analyzed by univariate or multivariate analysis., Results: The lymph node metastasis rate of stage T2 rectal cancer was significantly higher than that of stage T1[22.9% (108/463) vs. 9.7%(11/113), P=0.002], and the difference of stage T2a and T2b was not significant[22.0%(38/173) vs. 23.4% (68/290), P=0.733]. Multivariate analysis showed that poor differentiation(HR=1.54, 95% CI:1.12 to 2.13), abnormal carbohydrate antigen (CA) 199 level (HR=2.05, 95% CI:1.16 to 3.62), ulcerative mass (HR=1.58, 95% CI:1.05 to 2.39) and invasion of muscle (of inner ring muscle HR=3.55, 95% CI:1.79 to 7.02; of outer longitudinal muscle, HR=2.35, 95% CI:1.21 to 4.60) were independent risk factors of lymph node metastasis in patients with stage T1-T2 rectal cancer(all P<0.05). Meanwhile poor differentiation(HR=4.43, 95% CI:1.51 to 13.03), abnormal carcinoembryonic antigen(CEA) level (HR=4.66, 95% CI:1.18 to 20.11) and ulcerative mass (HR=6.23, 95% CI:1.51 to 25.66) were risk factors of lymph node metastasis in patients with stage T1 rectal cancer., Conclusion: Poor differentiation, preoperative high CA199, ulcerated tumor, invasion of inner ring muscle or outer longitudinal muscle are risk factors of lymph node metastasis in patients with stage T1-T2 rectal cancer, while the invasion depth of muscularis propria is not risk factor. Besides, poor differentiation, abnormal CEA level, ulcerated tumor are risk factors of lymph node metastasis in stage T1 rectal cancer patients, which can be used as reference for local excision in patients with stage T1 rectal cancer.

Objective: To compare the oncologic clearance and long-term outcomes between laparoscopic surgery and open surgery in radical resection of rectal cancer., Methods: Clinicopathological and follow-up data of 1184 cases with rectal cancer undergoing radical resection from July 2005 to December 2011 were analyzed retrospectively. According to the surgical method, cases were divided into laparoscopy group (104 cases) and open group(1080 cases). Demographics, number of harvested lymph nodes, distance between distal margin and tumor, incidence of anastomotic complications, disease-free survival (DFS) and overall survival(OS) were compared between the two groups., Results: There were no significant differences in the number of harvested lymph nodes (15.5 vs. 14.4, P>0.05), length of distal margin (2.5 cm vs. 2.1 cm, P>0.05) and incidence of anastomotic complications (1.9% vs. 1.9%, P>0.05) between the two groups. And there were no significant differences in DFS and OS between the two groups (both P>0.05). The 3-year and 5-year DFS in laparoscopy group were 79.0% and 69.3%, and were 78.0% and 72.5% in open group. The 3-year and 5-year OS in laparoscopy group were 93.5% and 81.2%, which were 87.6% and 80.7% in open group. There were no significant differences in DFS and OS after stratification by TNM stage., Conclusion: The oncologic clearance and long-term outcomes after laparoscopic surgery are comparable with open surgery in radical resection of rectal cancer.