As an important basis for lesion determination and diagnosis, medical image segmentation has become one of the most important and hot research fields in the biomedical field, among which medical image segmentation algorithms based on full convolutional neural network and U-Net neural network have attracted more and more attention by researchers. At present, there are few reports on the application of medical image segmentation algorithms in the diagnosis of rectal cancer, and the accuracy of the segmentation results of rectal cancer is not high. In this paper, a convolutional network model of encoding and decoding combined with image clipping and pre-processing is proposed. On the basis of U-Net, this model replaced the traditional convolution block with the residual block, which effectively avoided the problem of gradient disappearance. In addition, the image enlargement method is also used to improve the generalization ability of the model. The test results on the data set provided by the "Teddy Cup" Data Mining Challenge showed that the residual block-based improved U-Net model proposed in this paper, combined with image clipping and preprocessing, could greatly improve the segmentation accuracy of rectal cancer, and the Dice coefficient obtained reached 0.97 on the verification set.
Algorithms ; Delayed Emergence from Anesthesia ; Humans ; Image Processing, Computer-Assisted ; Rectal Neoplasms/diagnostic imaging* ; Tomography, X-Ray Computed

Objective: To explore the application value of diffusion kurtosis imaging (DKI) in the differential diagnosis of rectal tumors and evaluating the prognostic factors associated with rectal adenocarcinoma. Methods: A total of 105 patients with rectal tumors admitted in the First Affiliated Hospital of Zhengzhou University from December 2018 to August 2020 were retrospectively analyzed. All patients underwent high-resolution magnetic resonance DKI scanning. The mean diffusivity (MD), mean kurtosis (MK) and apparent diffusion coefficient (ADC) were measured and the relationship of these parameters with pathological types and prognostic factors of rectal tumor were analyzed. The diagnostic efficacy of MD, MK, and ADC for positive circumferential resection margin (CRM) and extramural venous invasion (EMVI) of rectal adenocarcinoma was evaluated by the receiver operating characteristic (ROC) curve. Results: MD and ADC were only related to pathological type. The MD and ADC were (2.091±0.390)×10(-3) and (1.478±0.265)×10(-3) mm(2)/s in mucinous adenocarcinoma, higher than (1.136±0.182)×10(-3) and (0.767±0.077)×10(-3) mm(2)/s in unspecified adenocarcinoma and (1.617±0.697)×10(-3) and (0.940±0.179)×10(-3) mm(2)/s in tubulo-villous adenoma. The MD and ADC in unspecified adenocarcinoma were lower than those in tubule-villous adenoma (P<0.05). Nevertheless, MK was associated with pathological type, N stage, CRM and EMVI. The MK was 0.566±0.110 in mucinous adenocarcinoma, lower than 0.982±0.135 in unspecified adenocarcinoma and 0.827±0.121 in tubulo-villous adenoma. The MK in unspecified adenocarcinoma was higher than that in intubulo-villous adenoma. The MK was 0.984±0.107 in pN1-2, higher than 0.881±0.146 in pN0. The MK was 0.990±0.142 in positive CRM, higher than 0.862±0.114 in negative CRM. The MK was 0.996±0.140 in positive EMVI, higher than 0.832±0.100 in negative EMVI (P<0.05). The ROC curves showed that the AUCs of MD, MK and ADC in diagnosing positive CRM were 0.459, 0.653 and 0.408, respectively; with MK=1.006 as the optimal diagnostic threshold, the diagnostic sensitivity and specificity were 51.9% and 81.0%, respectively. The AUCs of MD, MK and ADC values in diagnosing positive EMVI were 0.510, 0.662 and 0.388, respectively; with MK=1.010 as the optimal diagnostic threshold, the diagnostic sensitivity and specificity were 50.9% and 87.5%, respectively. Conclusions: DKI quantitative parameter is helpful for discriminating rectal tubulo-villous adenoma, unspecified adenocarcinoma, and mucinous adenocarcinoma, and is helpful for predicting the prognosis of patients with rectal adenocarcinoma. High MK is associated with positive CRM and EMVI.
Humans ; Adenocarcinoma/diagnostic imaging* ; Adenocarcinoma, Mucinous/diagnostic imaging* ; Adenoma, Villous/diagnostic imaging* ; Diffusion Magnetic Resonance Imaging/methods* ; Prognosis ; Rectal Neoplasms/pathology* ; Retrospective Studies ; Sensitivity and Specificity

At present, preoperative assessment of rectal cancer stage mainly relies on imaging examination, and the results of imaging reading will directly determine the treatment. In order to alleviate the reading pressure of the radiologist and improve the efficiency and accuracy of imaging diagnosis, there are related studies on using artificial intelligence automatic recognition system to assist the imaging assessment of rectal cancer staging now. Colorectal Surgery Group, Chinese Society of Surgery of the Chinese Medical Association, along with Beihang University, proposed the expert's advice on the surgical clinical application of rectal cancer staging recognition system based on artificial intelligence platform, so as to guide the standard application of this technology and promote the automation and intelligence of imaging reading.
Artificial Intelligence ; Diagnostic Imaging ; Digestive System Surgical Procedures ; Humans ; Magnetic Resonance Imaging ; Neoplasm Staging ; Rectal Neoplasms/surgery*

BACKGROUND: An artificial intelligence system of Faster Region-based Convolutional Neural Network (Faster R-CNN) is newly developed for the diagnosis of metastatic lymph node (LN) in rectal cancer patients. The primary objective of this study was to comprehensively verify its accuracy in clinical use. METHODS: Four hundred fourteen patients with rectal cancer discharged between January 2013 and March 2015 were collected from 6 clinical centers, and the magnetic resonance imaging data for pelvic metastatic LNs of each patient was identified by Faster R-CNN. Faster R-CNN based diagnoses were compared with radiologist based diagnoses and pathologist based diagnoses for methodological verification, using correlation analyses and consistency check. For clinical verification, the patients were retrospectively followed up by telephone for 36 months, with post-operative recurrence of rectal cancer as a clinical outcome; recurrence-free survivals of the patients were compared among different diagnostic groups, by methods of Kaplan-Meier and Cox hazards regression model. RESULTS: Significant correlations were observed between any 2 factors among the numbers of metastatic LNs separately diagnosed by radiologists, Faster R-CNN and pathologists, as evidenced by rradiologist-Faster R-CNN of 0.912, rPathologist-radiologist of 0.134, and rPathologist-Faster R-CNN of 0.448 respectively. The value of kappa coefficient in N staging between Faster R-CNN and pathologists was 0.573, and this value between radiologists and pathologists was 0.473. The 3 groups of Faster R-CNN, radiologists and pathologists showed no significant differences in the recurrence-free survival time for stage N0 and N1 patients, but significant differences were found for stage N2 patients. CONCLUSION: Faster R-CNN surpasses radiologists in the evaluation of pelvic metastatic LNs of rectal cancer, but is not on par with pathologists. TRIAL REGISTRATION www.chictr.org.cn (No. ChiCTR-DDD-17013842).
Adult ; Aged ; Aged, 80 and over ; Artificial Intelligence ; Female ; Humans ; Lymphatic Metastasis ; Male ; Middle Aged ; Neoplasm Recurrence, Local ; Neoplasm Staging ; Neural Networks (Computer) ; Pathologists ; Radiologists ; Rectal Neoplasms ; diagnostic imaging ; mortality ; pathology

Neoadjuvant chemoradiotherapy (NCRT) combined with total mesorectal excision (TME) has become the recommended standard treatment strategy for local advanced rectal cancer (cT3 or cN+). After neoadjuvant chemoradiotherapy, preoperative T/N re-staging and efficacy evaluation of rectal cancer are directly related to the available treatment options and prognosis, so they are the common questions concerned by physicians. At present, magnetic resonance imaging (MRI) is acknowledged to be one of the more effective and feasible methods of T/N re-staging and efficacy evaluation, especially in the molecular microscopic scale. The diffusion weighted imaging (DWI) can reflect the movement of water molecules outside the tumor cells, and the multi-phase dynamic contrast enhanced MRI (DCE-MRI) can indirectly reflect the permeability of tumor vascular wall and local blood perfusion of tumor from the view of pathophysiological point. Because of the influence of edema, inflammatory response and fibrous tissue proliferation after radiotherapy, scholars both at home and abroad increasingly pay more attentions to the accuracy of T/N re-staging and efficacy prediction in MRI following neoadjuvant therapy. In this review, we elucidate the application value and limitation of MRI based on T/N re-staging and local efficacy evaluation.
Chemoradiotherapy ; Humans ; Magnetic Resonance Imaging ; Neoadjuvant Therapy ; Neoplasm Staging ; Rectal Neoplasms ; diagnostic imaging ; therapy ; Treatment Outcome

OBJECTIVETo explore the application value of texture analysis of magnetic resonance images (MRI) in predicting the efficacy of neoadjuvant chemoradiotherapy(nCRT) for rectal cancer.

METHODSA total of 34 rectal cancer patients who were hospitalized at Zhejiang Provincial People's Hospital from February 2015 to April 2017 were prospectively enrolled and received 3.0T MRI examination at pre-nCRT (1 day before nCRT), early stage (at 10-day after nCRT) and middle stage (at 20-day after nCRT).

INCLUSION CRITERIAdistance from tumor lower margin to anal edge was less than 12 cm under rectoscope; rectal cancer was confirmed by preoperative pathology; clinical stage was T3 or above; lymph node metastasis existed but without distant metastasis; functions of liver, kidney and heart present no contraindications of operation.

EXCLUSION CRITERIAunfinished nCRT, surgery and three examinations of MRI; image motion artifacts; lack of postoperative pathological results. All the patients underwent rectal cancer long-term three-dimensional radiotherapy and chemotherapy combined with nCRT (oxaliplatin plus capecitabine). The tumor regression grading (TRG) was divided into TRG 0 to 4 grade after nCRT, and TRG 4 was classified as pathological complete remission (pCR); TRG 2 to 3 was classified as partial remission (PR); the rest was no remission (NR). Extraction and analysis of texture features in T2-weighted MR-defined tumor region were performed using Omni Kinetics texture software. The texture values of each time point were statistically analyzed, and the differences of texture values and change differences between pCR and PR+NR, and NR and pCR+PR were compared respectively. Statistically significant texture values were screened and were used in receiver operating characteristic (ROC) curve to assess the prediction of the efficacy of nCRT.

RESULTSOf 34 patients, 21 were males and 13 were females with median age of 49.3 years. Nineteen (55.9%) patients were low rectal adenocarcinoma and 15 (44.1%) patients were middle rectal adenocarcinoma. Nine (26.5%) cases belonged to pCR, 13 (38.2%) belonged to PR, and 12 (35.3%) belonged to NR. Before nCRT, the entropy of tumor area in pCR patients was significantly higher than that in PR+NR patients (7.164±0.272 vs. 6.823±0.309, t=2.925, P=0.006). At the middle stage of nCRT, as compared with PR+NR patients for the texture features of tumor region, the variance (1566±281 vs. 2883±867, t=-4.435, P=0.000) and entropy(5.436±0.934 vs. 6.803±0.577, t=-4.118,P=0.002) of pCR patients were significantly lower; kurtosis(4.800±1.288 vs. 3.206±1.211, t=3.333, P=0.002) and energy (0.016±0.005 vs. 0.010±0.004, t=3.240, P=0.003) of pCR patients were significantly higher. As compared to pCR+PR patients, the kurtosis(2.461±0.931 vs. 4.264±1.205, t=-4.493, P=0.000) and energy (0.011±0.004 vs. 0.014±0.004, t=-3.453, P=0.000) of the NR patients were significantly lower. As for texture change values between early stage and middle stage, the entropy difference was significant between pCR and PR+NR, NR and pCR+PR (1.344±0.819 vs. 0.489±0.319, t=3.047, P=0.014; 0.446±0.213 vs. 0.917±0.677, t=-3.638, P=0.001, respectively). As for texture change values between pre-nCRT and middle stage, variance and entropy differences between pCR and PR+NR (1759±1226 vs. 977±842, t=2.113, P=0.042; 1.728±0.918 vs. 0.524±0.355, t=3.832, P=0.004), and the change values of entropy between NR and pCR+PR (0.475±0.349 vs. 1.044±0.860, t=-2.722, P=0.011) were statistically significant. The above indicators were included in the ROC curve. The results revealed that at the middle stage, entropy value >5.983 indicated the best efficacy for the diagnosis of pCR, with the area under the ROC curve (AUC) of 0.885, the sensitivity of 100%, and the specificity of 66.7%; the energy <0.010 indicated the best AUC for diagnosis of NR was 0.902, with the sensitivity of 91.7% and specificity of 81.8%.

CONCLUSIONSTexture analysis based on T2 weighted images can predict the efficacy of nCRT for rectal cancer. The middle stage of nCRT is the best time of prediction. The entropy and energy of this period are texture parameters having higher predictive ability.

Chemoradiotherapy ; Female ; Humans ; Magnetic Resonance Spectroscopy ; Male ; Middle Aged ; Neoadjuvant Therapy ; Neoplasm Staging ; Predictive Value of Tests ; Prognosis ; Rectal Neoplasms ; diagnostic imaging ; therapy ; Treatment Outcome

OBJECTIVETo evaluate the diagnostic value of 3.0T high-resolution MRI in mesorectal lymph node metastasis of rectal cancer.

METHODSThe images and postoperative pathological data of patients with pathologically diagnosed rectal cancer who underwent prospective 3.0T two dimensional high-resolution MRI rectal examinations and surgery within two weeks after MRI examination at the First Affiliated Hospital, Sun Yat-sen University from November 2015 to November 2016 were retrospectively collected. Patients who received preoperative neoadjuvant therapy and those who did not undergo operation after MRI examination were excluded. The MRI sequences included high-resolution sagittal, coronal and oblique axial T2 weighted image (T2WI) (repetition time/echo time, 3000-4000 ms/77-87 ms; slice thickness/gap, 3 mm/0 mm; field of view, 18-22 cm). Two abdominal MRI radiologists independently assessed the morphology, margin, signal of all visible mesorectal nodes, measured their minor axes (three times for each radiologist) and gave estimation of the malignancy. The criteria of metastatic nodes on high-resolution MRI T2WI were nodes with irregular shape, ill-defined border and/or heterogeneous signal. The results of MRI diagnosis were compared with postoperative pathology. The sensitivity, specificity, accuracy, positive predictive value(PPV) and negative predictive value(NPV) of mesorectal nodes and nodes with different short-axis diameter ranges were calculated to evaluate the diagnostic efficiency of high-resolution MRI. Kappa statistics was used to evaluate the agreement for per node and for per patient between high-resolution MRI and pathological results. A Kappa value of 0-0.20 indicated poor agreement; 0.21-0.40 fair agreement; 0.41-0.60 moderate agreement; 0.61-0.80 good agreement; and 0.81-1.00 excellent agreement.

RESULTSA total of 81 patients were enrolled in the retrospective cohort study, including 50 males and 31 females with age of (59.3±11.1) years. Histopathology showed 1 case of well differentiated adenocarcinoma, 63 of moderately differentiated adenocarcinoma, 9 of moderately to poorly differentiated adenocarcinoma, 2 of poorly differentiated adenocarcinoma, 3 of mucinous adenocarcinoma and 3 of tubulovillous adenocarcinoma. Histopathological staging showed 2 cases in T1 stage, 20 in T2 stage, 45 in T3 stage and 14 in T4 stage; 34 in N0 stage, 40 in N1 stage and 7 in N2 stage; 76 in M0 stage and 5 in M1 stage. A total of 377 nodes were included in the node-by-node evaluation, of which 168 (44.6%) nodes were metastatic from 58.0% (47/81) patients. The median short-axis diameter was 5.4(2.4-18.6) mm in metastatic nodes, which was significantly larger than 3.8 (2.0-8.7) mm in non-metastatic nodes[Z=10.586, P=0.000]. The sensitivity, specificity, accuracy, PPV and NPV were 74.4% (125/168), 94.7% (198/209), 85.7% (323/377), 91.9% (125/136) and 82.2% (198/241), respectively. The Kappa values between high-resolution MRI and histopathological diagnosis for node-by-node and patient-by-patient were 0.71 and 0.70 respectively, indicating good agreements. Fourteen nodes >10 mm were all metastatic. The results of high-resolution MRI for nodal status were consistent with the results of histopathological diagnosis, and the sensitivity, accuracy and PPV were all 100.0%. Among 124 nodes with short-axis diameter of 5-10 mm, 95 (76.6%) were metastatic, and the sensitivity, specificity, accuracy, PPV and NPV were 78.9% (75/95), 86.2% (25/29), 80.6% (100/124), 94.9% (75/79) and 55.6% (25/45), respectively. The agreement was fair (Kappa value 0.55) between high-resolution MRI and histopathological diagnosis. Among 239 nodes with short-axis diameter ≤5 mm, 59(24.7%) were metastatic, and the sensitivity, specificity, accuracy, PPV and NPV were 61.0% (36/59), 96.1%(173/180), 87.4%(209/239), 83.7%(36/43) and 88.3%(173/196), respectively. The agreement was good (Kappa value 0.63) between high-resolution MRI and histopathological diagnosis.

CONCLUSIONRectal high-resolution MRI has good diagnostic value for estimating metastatic mesorectal nodes by evaluating the morphology, margin and signal of nodes.

Aged ; Female ; Humans ; Lymph Nodes ; Lymphatic Metastasis ; diagnostic imaging ; Magnetic Resonance Imaging ; Male ; Middle Aged ; Neoplasm Staging ; Prospective Studies ; Rectal Neoplasms ; diagnostic imaging ; pathology ; Retrospective Studies ; Sensitivity and Specificity

OBJECTIVETo explore the factors affecting the operative difficulty of triple-port laparoscopic surgery (TLS) in anterior resection.

METHODSA retrospective case-control study was carried out. Clinical and MRI imaging data of 106 colorectal cancer cases undergoing TLS anterior resection at Department of Colorectal Surgery of Ruijin Hospital between 2013 and 2016 were retrospectively analyzed.

INCLUSION CRITERIA(1) patients receiving TLS anterior resection (Dixon operation); (2) preoperative stageI( to III( malignant tumor;(3) distance of 5-15 cm from inferior margin of tumor to anal verge; and (4) available preoperative rectal MRI.

EXCLUSION CRITERIA(1) patients receiving preoperative adjuvant therapy; (2) patients with low rectal cancer or with local advanced disease; (3) T4b tumor. Rectal MRI was introduced to measure the structure of pelvis. In sagittal view, superior margin of the first sacral vertebrae, superior margin of the third sacral vertebrae, apex of coccyx, and the line of superior margin of pubic symphysis were used to form a pentagon. The 5 lines were marked as N, O, P, Q, R, and the 5 included angles were marked as angle 1, 2, 3, 4, 5. Organs (uterus and prostate) and tumor (transverse diameter, longitudinal diameter, section area, lesion length, distance to circumference cutting edge) were also measured on MRI. The operative time was applied to be the indicator of operative difficulty and patients were divided into 2 groups according to median operative time. Baseline information (age, gender, BMI, distance from inferior margin of tumor to anal verge, operative history, length of tumor), preoperative tumor staging, and MRI measurements (pelvis, tumor, uterus, prostate), etc were compared between two groups. Factors affecting operative difficulty of TLS were analyzed with logistic regression model.

RESULTSOf 106 enrolled patients, 73 were male and 33 female with mean age of (59.8±12.2) years and mean BMI of (22.8±3.3) kg/m; 25 patients had previous abdominal surgery; distance from inferior margin of tumor to anal verge was (7.4±2.0) cm and the tumor diameter was (3.7±1.4) cm; 24, 36 and 46 patients were in stage I(, II( and III( respectively. All operations were completed successfully. The median number of harvested lymph node was 13(11-16); the median length of distal resection margin was 2.5(2.0-3.1) cm; the median operative time was 2.0(1.5-2.6) hours; the median intraoperative blood loss was 50(0-100) ml; the median time to liquid diet was 4(3-5) days; the median hospital stay was 7(6-10) days. Ten cases (9.4%) developed complications within 30 days after surgery. Patients were divided into ≤2 h group and > 2 h group according to median operative time, and both groups had 53 patients. As compared to ≤2 h group, >2 h group had shorter distance from inferior margin of tumor to anal verge [(6.8 ± 1.5) cm vs. (8.0 ± 2.4) cm, t = 3.174, P = 0.004], lower ratio of (R+N)/(O+P)(1.61±0.27 vs. 1.73±0.19, t = 2.494, P = 0.014), larger transverse distance of tumor [(3.45±0.72) cm vs. (3.05±0.89) cm, t = 0.224, P = 0.027]. Multivariate logistic regression analysis showed the distance from inferior margin of tumor to anal verge was the independent factor affecting operative difficulty(OR=0.584, 95%CI:0.429-0.796, P = 0.001).

CONCLUSIONSSurgeons may have less difficulty in performing TLS anterior resection for patients with longer distance from inferior margin of tumor to anal verge. In preoperative assessment of operative difficulty of TLS, comprehensive evaluation should be performed. Distance from inferior margin of tumor to anal verge should be regarded as the main factor, and MRI (R+N)/(O+P) and transverse diameter of tumor should be used as important reference, leading to reasonable choice of cases for TLS and smooth pass of study curve.

Aged ; Anal Canal ; Case-Control Studies ; Female ; Humans ; Laparoscopy ; methods ; Male ; Middle Aged ; Rectal Neoplasms ; diagnostic imaging ; surgery ; Retrospective Studies ; Treatment Outcome

This study was performed to evaluate prostate-specific antigen-age volume (PSA-AV) scores in predicting prostate cancer (PCa) in a Chinese biopsy population. A total of 2355 men who underwent initial prostate biopsy from January 2006 to November 2015 in Huashan Hospital were recruited in the current study. The PSA-AV scores were calculated and assessed together with PSA and PSA density (PSAD) retrospectively. Among 2133 patients included in the analysis, 947 (44.4%) were diagnosed with PCa. The mean age, PSA, and positive rates of digital rectal examination result and transrectal ultrasound result were statistically higher in men diagnosed with PCa (all P < 0.05). The values of area under the receiver operating characteristic curves (AUCs) of PSAD and PSA-AV were 0.864 and 0.851, respectively, in predicting PCa in the entire population, both performed better than PSA (AUC = 0.805; P < 0.05). The superiority of PSAD and PSA-AV was more obvious in subgroup with PSA ranging from 2.0 ng ml^-1 to 20.0 ng ml^-1. A PSA-AV score of 400 had a sensitivity and specificity of 93.7% and 40.0%, respectively. In conclusion, the PSA-AV score performed equally with PSAD and was better than PSA in predicting PCa. This indicated that PSA-AV score could be a useful tool for predicting PCa in Chinese population.
Aged ; Aged, 80 and over ; Aging/pathology* ; Area Under Curve ; Asian People ; Digital Rectal Examination ; Humans ; Image-Guided Biopsy ; Male ; Predictive Value of Tests ; Prostate-Specific Antigen/blood* ; Prostatic Neoplasms/diagnostic imaging* ; ROC Curve ; Reference Values ; Retrospective Studies ; Sensitivity and Specificity ; Ultrasonography, Interventional

Rectal cancer is one of the common cancers which poses a threat to the health of mankind. In recent years. Multi-modality treatment strategies for locally advanced rectal cancer improve the treatment efficiency. Accurate prediction of the treatment response after the neoadjuvant chemoradiotherapy (CRT) can guide more suitable treatment strategy. MERCURY study proved the prognostic value of post-CRT standard morphologic MRI(T2-weighted) assessment of tumor regression grade(TRG), and MRI assessment of circumferential resection margin can guide the definitive surgery. Compared with standard morphologic MRI (T2-weighted), functional MRI, including diffusion weighted imaging (DWI) and dynamic contrast enhanced (DCE) MRI, has shown more promising results for the prediction of therapeutic response in rectal cancer. The addition of diffusion-weighted images to T2-weighted images improves the accuracy of restaging examinations for determination of complete pathologic responders. DCE can reflect the tumor micro-vascular environment, and the change of perfusion in response to treatment. These images have the potential to improve the accuracy of therapeutic response in rectal cancer.
Chemoradiotherapy ; statistics & numerical data ; Contrast Media ; Diffusion Magnetic Resonance Imaging ; statistics & numerical data ; Humans ; Magnetic Resonance Imaging ; methods ; statistics & numerical data ; Margins of Excision ; Neoadjuvant Therapy ; statistics & numerical data ; Neoplasm Staging ; instrumentation ; methods ; statistics & numerical data ; Prognosis ; Rectal Neoplasms ; blood supply ; diagnostic imaging ; pathology