Objective To propose a lung artery segmentation method that integrates shape and position prior knowledge, aiming to solve the issues of inaccurate segmentation caused by the high similarity and small size differences between the lung arteries and surrounding tissues in CT images. Methods Based on the three-dimensional U-Net network architecture and relying on the PARSE 2022 database image data, shape and position prior knowledge was introduced to design feature extraction and fusion strategies to enhance the ability of lung artery segmentation. The data of the patients were divided into three groups: a training set, a validation set, and a test set. The performance metrics for evaluating the model included Dice Similarity Coefficient (DSC), sensitivity, accuracy, and Hausdorff distance (HD95). Results The study included lung artery imaging data from 203 patients, including 100 patients in the training set, 30 patients in the validation set, and 73 patients in the test set. Through the backbone network, a rough segmentation of the lung arteries was performed to obtain a complete vascular structure; the branch network integrating shape and position information was used to extract features of small pulmonary arteries, reducing interference from the pulmonary artery trunk and left and right pulmonary arteries. Experimental results showed that the segmentation model based on shape and position prior knowledge had a higher DSC (82.81%±3.20% vs. 80.47%±3.17% vs. 80.36%±3.43%), sensitivity (85.30%±8.04% vs. 80.95%±6.89% vs. 82.82%±7.29%), and accuracy (81.63%±7.53% vs. 81.19%±8.35% vs. 79.36%±8.98%) compared to traditional three-dimensional U-Net and V-Net methods. HD95 could reach (9.52±4.29) mm, which was 6.05 mm shorter than traditional methods, showing excellent performance in segmentation boundaries. Conclusion The lung artery segmentation method based on shape and position prior knowledge can achieve precise segmentation of lung artery vessels and has potential application value in tasks such as bronchoscopy or percutaneous puncture surgery navigation.

Objective:To investigate the impact of relative locations of multiple foci and microsatellite status of sporadic, synchronous, multiple, primary, colorectal carcinomas on clinicopathological features and prognosis.Methods:The clinicopathologic and prognostic data of 278 patients with sporadic, synchronous, multiple, primary, colorectal carcinomas who had been admitted to the Department of Colorectal Surgery at Zhejiang Cancer Hospital from January 2008 to July 2022 were retrospectively collected. The patients were categorized into three groups based on the relative locations of their multiple cancer foci: (1) a right-sided group that comprised patients with multiple cancer foci in the cecum, ascending colon, hepatic flexure of the colon, and transverse colon; (2) a left-sided group that comprised patients with multiple cancer foci in the splenic flexure of the colon, descending colon, sigmoid colon, and rectum; and (3) a left- and right-sided group that comprised patients with multiple cancer foci in the right half of the colon and left half of the colon/rectum. Additionally, the patients were further divided into two groups based on microsatellite status: a high microsatellite instability (MSI-H) and a low MSI/stable MSI (MSI/L&MSS) group. We compared differences in clinical characteristics and prognostic indicators between these groups. The χ 2 test was utilized to compare selected clinical characteristics, whereas Kaplan-Meier survival analyses and log-rank tests were performed to compare their effects on prognosis. Result:Among 278 patients with SSCRC, 256 (92.1%) presented with two cancer foci and 22 (7.9%) with more than two foci. Additionally, 255 patients (91.7%) had adenocarcinomas, whereas the remaining 23 (8.3%) had mucinous adenocarcinomas. Lymph node metastases were identified in 136 patients (48.9%); the cancer foci had infiltrated beyond the muscular layer in 238 (85.6%); and 147 patients (52.9%) were diagnosed with TNM Stage III–IV disease. There were 155 patients (55.8%) in the left-sided group, 55 (19.8%) in the right-sided group, and 68 (24.5%) in the left- and right-sided group. Immunohistochemical examination of all four mismatch repair proteins were performed in 199 cases, revealing that 166 of these patients had MSI/L&MSS and 33 MSI-H disease. In the left-sided, left- and right-sided, and right-sided groups, the proportion of women was 16.8% (26/155), 26.5% (18/68), and 49.1% (27/55), respectively; these differences are statistically significant (χ 2=22.335, P<0.001). The proportions of patients with more than three cancer foci were 5.2% (8/155), 16.2% (11/68), and 5.5% (3/55), respectively; these differences are statistically significant (χ 2=8.438, P=0.015). The proportions of mucinous adenocarcinomas were 4.5% (7/155), 8.8% (6/68), and 18.2% (10/55), respectively; these differences are statistically significant (χ 2=10.026, P=0.007). The proportions of patients with lymph node metastases were 55.5% (86/155), 48.5% (33/68), and 30.9% (17/55); these differences are statistically significant (χ 2=9.817, P=0.007). The proportions of patients with Stage T3 & T4 disease in each group according to location were 81.3% (126/155), 88.2% (60/68), and 94.5% (52/55), respectively; these differences are statistically significant (χ 2=6.293, P=0.043). The proportions of TNM Stage III–IV tumors were 59.4% (92/155), 54.4% (37/68), and 32.7% (18/55), respectively; these differences are statistically significant (χ 2=11.637, P=0.003). Age, size of cancer foci, presence of distant metastasis, adenoma, nerve invasion, and vascular invasion did not differ significantly between the three groups (all P>0.05). Compared with those with MSI-H, patients with MSI/L&MSS disease were more likely to be aged >65 years and male (50.6% [84/166] vs. 15.2% [5/33], χ 2=13.994, P<0.001; 80.7% [134/166] vs. 54.5% [18/33], χ 2=10.457, P=0.001), more likely to be in the left-sided group (63.3% [105/166] vs. 24.2% [8/33], χ 2=18.232, P<0.001), had a higher proportion of cancer foci of diameter <4 cm (54.8% [91/166] vs. 33.3% [11/33], χ 2=5.086, P=0.024), and a lower proportion of mucinous adenocarcinomas (4.2% [7/166] vs. 27.3% [9/33], χ 2=19.791, P<0.001), more likely to develop distant metastases (22.3% [37/166] vs. 6.1% [2/33], χ 2=4.601, P=0.032), more likely to have lymph node metastases (57.2% [95/166) vs. 24.2% [8/33], χ 2=11.996, P<0.001) and nerve invasion (28.9% [48/166] vs. 6.1% [2/33], χ 2=7.643, P=0.006), had a higher proportion of TNM Stage III–IV disease (60.2% [100/166] vs. 24.2% [8/33], χ 2=14.374, P<0.001), and a smaller proportion of family history of tumors (28.9% [48/166] vs. 60.6% [20/33], χ 2=12.228, P<0.001). All the above-listed differences are statistically significant (all P<0.05). The differences in number of cancer foci, depth of infiltration, presence or absence of adenomas, and vascular invasion were not statistically significant (all P>0.05). In the 33 patients with MSI-H status and mismatch repair protein loss, the highest frequency of deletion was found in PMS-2 (66.7%, 22/33), followed by MLH-1 (57.6%, 19/33), whereas the proportions of MSH-2 (33.3%, 11/33) and MSH-6 (24.2%, 8/33) deletions were relatively low. There were statistically significant differences in the 3-year overall survival rates among the groups according to relative locations of cancer foci. The 3-year overall survival rates were 96.8%, 79.6%, and 88.5% in the right-sided, left- and right-sided, and left-sided groups, respectively ( P=0.021). As to microsatellite status, the 3-year overall survival rate of patients with MSI-H disease was 93.8%, which is significantly better than the 78.4% for those with MSI/L & MSS ( P=0.026). Conclusions:Among sporadic, synchronous, multiple, primary, colorectal carcinomas, those with right-sided disease had the deepest local infiltration, whereas those with left-sided disease had the greatest number of lymph node metastases, most advanced clinical TNM stage, lowest percentage of MSI-H disease, and the poorest prognosis.

Objective:To investigate the impact of relative locations of multiple foci and microsatellite status of sporadic, synchronous, multiple, primary, colorectal carcinomas on clinicopathological features and prognosis.Methods:The clinicopathologic and prognostic data of 278 patients with sporadic, synchronous, multiple, primary, colorectal carcinomas who had been admitted to the Department of Colorectal Surgery at Zhejiang Cancer Hospital from January 2008 to July 2022 were retrospectively collected. The patients were categorized into three groups based on the relative locations of their multiple cancer foci: (1) a right-sided group that comprised patients with multiple cancer foci in the cecum, ascending colon, hepatic flexure of the colon, and transverse colon; (2) a left-sided group that comprised patients with multiple cancer foci in the splenic flexure of the colon, descending colon, sigmoid colon, and rectum; and (3) a left- and right-sided group that comprised patients with multiple cancer foci in the right half of the colon and left half of the colon/rectum. Additionally, the patients were further divided into two groups based on microsatellite status: a high microsatellite instability (MSI-H) and a low MSI/stable MSI (MSI/L&MSS) group. We compared differences in clinical characteristics and prognostic indicators between these groups. The χ 2 test was utilized to compare selected clinical characteristics, whereas Kaplan-Meier survival analyses and log-rank tests were performed to compare their effects on prognosis. Result:Among 278 patients with SSCRC, 256 (92.1%) presented with two cancer foci and 22 (7.9%) with more than two foci. Additionally, 255 patients (91.7%) had adenocarcinomas, whereas the remaining 23 (8.3%) had mucinous adenocarcinomas. Lymph node metastases were identified in 136 patients (48.9%); the cancer foci had infiltrated beyond the muscular layer in 238 (85.6%); and 147 patients (52.9%) were diagnosed with TNM Stage III–IV disease. There were 155 patients (55.8%) in the left-sided group, 55 (19.8%) in the right-sided group, and 68 (24.5%) in the left- and right-sided group. Immunohistochemical examination of all four mismatch repair proteins were performed in 199 cases, revealing that 166 of these patients had MSI/L&MSS and 33 MSI-H disease. In the left-sided, left- and right-sided, and right-sided groups, the proportion of women was 16.8% (26/155), 26.5% (18/68), and 49.1% (27/55), respectively; these differences are statistically significant (χ 2=22.335, P<0.001). The proportions of patients with more than three cancer foci were 5.2% (8/155), 16.2% (11/68), and 5.5% (3/55), respectively; these differences are statistically significant (χ 2=8.438, P=0.015). The proportions of mucinous adenocarcinomas were 4.5% (7/155), 8.8% (6/68), and 18.2% (10/55), respectively; these differences are statistically significant (χ 2=10.026, P=0.007). The proportions of patients with lymph node metastases were 55.5% (86/155), 48.5% (33/68), and 30.9% (17/55); these differences are statistically significant (χ 2=9.817, P=0.007). The proportions of patients with Stage T3 & T4 disease in each group according to location were 81.3% (126/155), 88.2% (60/68), and 94.5% (52/55), respectively; these differences are statistically significant (χ 2=6.293, P=0.043). The proportions of TNM Stage III–IV tumors were 59.4% (92/155), 54.4% (37/68), and 32.7% (18/55), respectively; these differences are statistically significant (χ 2=11.637, P=0.003). Age, size of cancer foci, presence of distant metastasis, adenoma, nerve invasion, and vascular invasion did not differ significantly between the three groups (all P>0.05). Compared with those with MSI-H, patients with MSI/L&MSS disease were more likely to be aged >65 years and male (50.6% [84/166] vs. 15.2% [5/33], χ 2=13.994, P<0.001; 80.7% [134/166] vs. 54.5% [18/33], χ 2=10.457, P=0.001), more likely to be in the left-sided group (63.3% [105/166] vs. 24.2% [8/33], χ 2=18.232, P<0.001), had a higher proportion of cancer foci of diameter <4 cm (54.8% [91/166] vs. 33.3% [11/33], χ 2=5.086, P=0.024), and a lower proportion of mucinous adenocarcinomas (4.2% [7/166] vs. 27.3% [9/33], χ 2=19.791, P<0.001), more likely to develop distant metastases (22.3% [37/166] vs. 6.1% [2/33], χ 2=4.601, P=0.032), more likely to have lymph node metastases (57.2% [95/166) vs. 24.2% [8/33], χ 2=11.996, P<0.001) and nerve invasion (28.9% [48/166] vs. 6.1% [2/33], χ 2=7.643, P=0.006), had a higher proportion of TNM Stage III–IV disease (60.2% [100/166] vs. 24.2% [8/33], χ 2=14.374, P<0.001), and a smaller proportion of family history of tumors (28.9% [48/166] vs. 60.6% [20/33], χ 2=12.228, P<0.001). All the above-listed differences are statistically significant (all P<0.05). The differences in number of cancer foci, depth of infiltration, presence or absence of adenomas, and vascular invasion were not statistically significant (all P>0.05). In the 33 patients with MSI-H status and mismatch repair protein loss, the highest frequency of deletion was found in PMS-2 (66.7%, 22/33), followed by MLH-1 (57.6%, 19/33), whereas the proportions of MSH-2 (33.3%, 11/33) and MSH-6 (24.2%, 8/33) deletions were relatively low. There were statistically significant differences in the 3-year overall survival rates among the groups according to relative locations of cancer foci. The 3-year overall survival rates were 96.8%, 79.6%, and 88.5% in the right-sided, left- and right-sided, and left-sided groups, respectively ( P=0.021). As to microsatellite status, the 3-year overall survival rate of patients with MSI-H disease was 93.8%, which is significantly better than the 78.4% for those with MSI/L & MSS ( P=0.026). Conclusions:Among sporadic, synchronous, multiple, primary, colorectal carcinomas, those with right-sided disease had the deepest local infiltration, whereas those with left-sided disease had the greatest number of lymph node metastases, most advanced clinical TNM stage, lowest percentage of MSI-H disease, and the poorest prognosis.

Recent advancement in natural language processing (NLP) and medical imaging empowers the wide applicability of deep learning models. These developments have increased not only data understanding, but also knowledge of state-of-the-art architectures and their real-world potentials. Medical imaging researchers have recognized the limitations of only targeting images, as well as the importance of integrating multimodal inputs into medical image analysis. The lack of comprehensive surveys of the current literature, however, impedes the progress of this domain. Existing research perspectives, as well as the architectures, tasks, datasets, and performance measures examined in the present literature, are reviewed in this work, and we also provide a brief description of possible future directions in the field, aiming to provide researchers and healthcare professionals with a detailed summary of existing academic research and to provide rational insights to facilitate future research.
Humans ; Natural Language Processing ; Surveys and Questionnaires

Objective    To explore the feasibility and clinical value of free-of-puncture positioning in three-dimension-guided anatomical segmentectomy for ground-glass nodule (GGN) compared with percutaneous positioning. Methods    Clinical data of 268 enrolled patients undergoing anatomical pulmonary segmentectomy from October 2018 to June 2019 were retrospectively collected, including 75 males and 193 females with an average age of 56.55±12.10 years. The patients were divided into two groups, including a percutaneous positioning group (n=89) and a free-of-puncture positioning group (n=179). Perioperative data of the two groups were compared. Results    The average CT scan times of the percutaneous positioning group was 3.01±0.98 times, and the numerical rating scale (NRS) score of puncture pain was 3.98±1.61 points. Pulmonary compression pneumothorax (≥30%) occurred in 7 (7.87%) patients and intercostal vascular hemorrhage occurred in 8 (8.99%) patients after puncture. Lung nodules were successfully found and removed in both groups. There was no statistically significant difference between the two groups in the location of nodules (P=0.466), operation time (151.83±39.23 min vs. 154.35±33.19 min, P=0.585), margin width (2.07±0.35 cm vs. 1.98±0.28 cm, P=0.750), or the number of excised subsegments (2.83±1.13 vs. 2.73±1.16, P=0.530). Conclusion    Anatomical segmentectomy with three-dimensional navigation avoids the adverse consequences of puncture, which has the same clinical efficacy and meets the requirements of oncology compared with percutaneous positioning. The free-of-puncture positioning method can be used for GGN located in the central region of pulmonary segment/subsegment or adjacent to   intersegment veins instead of percutaneous positioning.

BACKGROUND: For early-stage lung cancer, segmentectomy can get the same oncological benefits as lobectomy. Accurate identification of the intersegmental border is the key to segmentectomy. This study used extended segmentectomy and extended subsegmentectomy to treat lung intersegmental and intersubsegmental ground-glass nodules (GGN) by utilizing modified inflation-deflation methods to distinguish the intersegmental and intersubsegmental borders. The accuracy of modified inflation-deflation methods and the effectiveness of extended resection to guarantee a safe surgical margin were evaluated. METHODS: A retrospective analysis of 83 cases of extended segmentectomy and extended subsegmentectomy was conducted. Preoperative three-dimensional computed tomography bronchography and angiography (3D-CTBA) revealed that nodules were involved in intersegmental or intersubsegmental veins. Based on preoperative three-dimensional reconstruction, the surgery was designed to extendedly remove the dominant lung segment or subsegment with nodules involved. When the dominant lung segment or subsegment could not be identified, the simpler lung segment or subsegment was selected for the resection. After the target vessel and bronchus were cut off during the operation, modified inflation-deflation method was used to determine the border, and a stapler was used to resect the adjacent lung segment or subsegment tissue by 2 cm-3 cm around the inflation-deflation boundary line. Then, the relationship between the inflation-deflation boundary line and the nodule and the width of the surgical margin were measured. Clinical data were collected during the perioperative period. RESULTS: 56 extended segmentectomies and 27 extended subsegmentectomies were performed. The average diameter of pulmonary nodules was (0.9±0.3) cm. There were 79 cases with clearly inflation-deflation boundary lines. The average time needed for the appearance of the lines was (13.6±6.5) min. In 55 cases, the nodules were involved with the inflation-deflation boundary lines. Meanwhile, the remaining 24 cases revealed an average minimum distance of (0.6±0.3) cm between nodules and the boundary lines. The average width of surgical margin was (2.1±0.3) cm in these 79 cases. No deaths or major complications appeared during 30 d after operation. CONCLUSIONS The modified inflation-deflation method can effectively define the intersegmental and intersubsegmental borders, and guarantee the safe surgical margins of extended segmentectomy and extended subsegmentectomy to treat intersegmental and intersubsegmental small lung tumors.

Objective To investigate the establishment,operation and performance of external quality assessment(EQA) system for microbial morphology and detection of special drug-resistance in clinical laboratory,and explore the value of the developed system in clinical application.Methods The pictures of known bacteria and fungi colony,gram staining and acid-fast staining from clinical microbiology were distributed to the participating laboratories in Gansu province twice a year at regular intervals.The pictures of standard knowledge points from CLSI,such as special drug resistance were distributed simultaneously.All the participating laboratories were required to complete the interpretation for the pictures and report their resuhs in a scheduled time.Then the resuhs were summarized and analyzed as 3 modes:complete consistency,general consistency and non-consistency.Results During the 2 years when the EQA system for microbial morphology and detection of special drug-resistance were performed for 24 times,the rate of annual complete consistency increased year by year and reached to 91.3％ in 2015.Conclusion The EQA system based on the examinations of microbial morphology and CLSI standard knowledge points for clinical laboratory may supervise the staff of clinical microbiology laboratories in the hospitals at second grade or above to master the skills of morphological identification and learn CLSI knowledge points,so their professional skills of clinical microbiology could be comprehensively improved.

Objective: To study the urban and rural difference impacts of aging on health care expenditure. Methods: Using 2002-2011 panel data of 31 provinces throughout the country to build fixed effect models for the urban and rural areas and make comparative analysis. Results: For urban and rural areas, there are significant differences in the impact of aging on health care expenditure, the estimated coefficients are 0.13 and 0.48 respectively. Conclusion:Under the background of the aging process, it needs to strengthen the construction of rural medical insurance system in China, increase financial support and rural health resource allocation.

OBJECTIVEThe aim of this study was to compare the short-term outcomes for hand-assisted, laparoscopic, and open resection for rectal cancer.

METHODSThree hundred ninety patients with rectal cancer who underwent curative resection between June 2009 and June 2012 were included. Patients were classified into a hand-assisted group (HALS, n=101), a laparoscopic surgery group (LS, n=157), and an open surgery group (OS, n=132). Patient and disease characteristics, operative parameters, postoperative morbidity, pathological results and length of recovery were compared among three groups.

RESULTSThe mean operating time was (173±39) min for the HALS group, (231±61) min for the LS group, and (173±39) min for the OS group (P<0.01). Conversion rates did not differ between HALS and LS groups (2.0% vs 3.2%, P=0.708). The overall complication rates were 11.9%, 11.5%, and 19.7% in the HALS, LS and OS groups respectively (P=0.100). The specimen quality with a specimen length, distal resection margin, harvested lymph nodes, and positive lymph nodes did not differ among the three groups. Patients in the HALS and LS groups recovered significantly faster than those from the OS group.

CONCLUSIONSThis comparative study shows that HALS and LS can reproduce the equivalent short-term results of standard OS. HALS retained the minimal invasive advantages of LS, and significantly shorten the operation time.

Aged ; Female ; Humans ; Laparoscopy ; methods ; Laparotomy ; Male ; Middle Aged ; Rectal Neoplasms ; surgery ; Retrospective Studies ; Treatment Outcome

Objective The aim of this study was to compare the short-term outcomes for hand-assisted, laparoscopic, and open resection for rectal cancer. Methods Three hundred ninety patients with rectal cancer who underwent curative resection between June 2009 and June 2012 were included. Patients were classified into a hand-assisted group (HALS, n=101), a laparoscopic surgery group (LS, n=157), and an open surgery group (OS, n=132). Patient and disease characteristics, operative parameters, postoperative morbidity, pathological results and length of recovery were compared among three groups. Results The mean operating time was (173±39) min for the HALS group, (231±61) min for the LS group, and (173 ±39) min for the OS group (P<0.01). Conversion ratesdid not differ between HALS and LS groups (2.0% vs 3.2%, P=0.708). The overall complication rates were 11.9%, 11.5%, and 19.7% in the HALS, LS and OS groups respectively (P=0.100). The specimen quality with a specimen length, distal resection margin, harvested lymph nodes, and positive lymph nodes did not differ among the three groups. Patients in the HALS and LS groups recovered significantly faster than those from the OS group. Conclusions This comparative study shows that HALS and LS can reproduce the equivalent short-term results of standard OS. HALS retained the minimal invasive advantages of LS, and significantly shorten the operation time.