Evaluation of Muscle Mass and Malnutrition in Patients with Colorectal Cancer Using the Global Leadership Initiative on Malnutrition Criteria and Comparing Bioelectrical Impedance Analysis and Computed Tomography Measurements.

OBJECTIVES: The aim of this investigation was to evaluate the discrepancies between bioelectrical impedance analysis (BIA) and computed tomography (CT) in assessing skeletal muscle mass and identifying low muscle mass in patients with colorectal cancer. METHODS: This study recruited 137 patients with colorectal cancer from February 2028 to December 2023. CT scans were analyzed at the Lumbar 3 vertebral level to determine the area of skeletal muscle, which was then utilized to estimate whole-body skeletal muscle mass. [BIA] was also employed to measure skeletal muscle. Both skeletal muscle mass values [kg] were divided by height2 [m2] to calculate the skeletal muscle index [SMI, kg/m2], denoted as SMI-CT and SMI-BIA, respectively. RESULTS: The median age was 69.8 + 9.5 years, with the sex ratio being 88/49 [male/female]. Whereas more than one-third of the patients were classified as malnourished based on the Global Leadership Initiative on Malnutrition GLIM-CT criteria using L3-SMI [n = 36.5%], fewer patients were classified as malnourished based on GLIM-BIA using SMI-BIA [n = 19.0%]. According to the CT analysis [low SMI-L3], 52 [38.0%] patients were diagnosed as having poor muscle mass, whereas only 18 [13.1%] patients were identified as having low muscle mass using BIA [low SMIBIA]. The measured SMI showed a positive association with SMI-CT in all patients [r = 0.63, p < 0.001]. Using Bland-Altman evaluation, a significant mean bias of 0.45 + 1.41 kg/m2 [95% CI 0.21-0.70; p < 0.001] between SMI-BIA and SMI-CT was reported. Receiver operating characteristic (ROC) curves were generated to detect poor muscle mass using SMI-BIA with CT as the gold standard. The area under the curve (AUC) for SMI-BIA in identifying poor muscle mass was 0.714 (95% CI: 0.624-0.824), with a good cut-off value of 8.1 kg/m2, yielding a sensitivity of 68.3% and a specificity of 66.9%. CONCLUSIONS: BIA generally overestimates skeletal muscle mass in colorectal cancer patients when contrasted to CT. As a result, BIA may underestimate the prevalence of poor muscle mass and malnutrition according to the GLIM criteria in this patient population.

Does portal vein anatomy influence intrahepatic distribution of metastases from colorectal cancer?

BACKGROUND: Other than location of the primary colorectal cancer (CRC), a few factors are known to influence the intrahepatic distribution of colorectal cancer liver metastases (CRLM). We aimed to assess whether the anatomy of the portal vein (PV) could influence the intrahepatic distribution of CRLM. PATIENTS AND METHODS: Patients with CRLM diagnosed between January 2018 and December 2022 at two tertiary centers were included and imaging was reviewed by two radiologists independently. Intra-operator concordance was assessed according to the intraclass correlation coefficient (ICC). The influence of the diameter, angulation of the PV branches and their variations on the number and distribution of CRLM were compared using Mann-Whitney, Kruskal-Wallis, Pearson's Chi-square and Spearman's correlation tests. RESULTS: Two hundred patients were included. ICC was high (> 0.90, P < 0.001). Intrahepatic CRLM distribution was right-liver, left-liver unilateral and bilateral in 66 (33%), 24 (12%) and 110 patients (55%), respectively. Median number of CRLM was 3 (1-7). Type 1, 2 and 3 portal vein variations were observed in 156 (78%), 19 (9.5%) and 25 (12%) patients, respectively. CRLM unilateral or bilateral distribution was not influenced by PV anatomical variations (P = 0.13), diameter of the right (P = 0.90) or left (P = 0.50) PV branches, angulation of the right (P = 0.20) or left (P = 0.80) PV branches and was independent from primary tumor localisation (P = 0.60). No correlations were found between CRLM number and diameter (R: 0.093, P = 0.10) or angulation of the PV branches (R: 0.012, P = 0.83). CONCLUSIONS: PV anatomy does not seem to influence the distribution and number of CRLM.

Role of Intravoxel Incoherent Motion and Dynamic-Contrast-Enhanced Perfusion in Response Prediction of Patients with Colorectal Cancer after Neoadjuvant Chemotherapy.

BACKGROUND: Colorectal cancer (CRC) is a complex malignancy requiring multimodal treatment strategies, including neoadjuvant chemoradiation therapy (Neo-CRT), to improve patient outcomes. However, the response to Neo-CRT varies among individuals, which necessitates the development of reliable predictors of treatment response. The present study aimed to investigate the role of intravoxel incoherent motion (IVIM) and dynamic contrast-enhanced (DCE) perfusion in predicting treatment response in CRC patients after Neo-CRT. METHODS: This study was conducted on patients diagnosed with locally advanced CRC who received Neo-CRT. IVIM and DCE perfusion imaging were performed before and after CRT. Quantitative parameters, including perfusion fraction (f), diffusion coefficient (D), and transfer constant (Ktrans), were calculated from the imaging data. Treatment response was assessed based on the pathological response after surgery. Statistical data were analyzed in SPSS v. 26 using the t-test and the chi-square method. RESULTS: A total of 51 patients (female: male = 22:29, mean age = 58.14±3.49) participated in the study. Among all the patients, 15 (29.4%) cases had good responses, while 36 (70.58%) cases did not respond to treatment. All DCE parameters showed higher sensitivity and specificity than IVIM D*. Ve, Kep, and DCE Ktrans indicated significant predictive power for treatment response. Ktrans was the most accurate parameter for predicting response to treatment. Overall sensitivity and specificity of DCE were 88.8% [95% CI: 80-95.6], and 80 % [95% CI: 65-90], and those of IVIM were 65.5% and 81%, respectively. Sensitivity and specificity for DCE + IVIM were 79.5% and 93.5%, and those of DCE + IVIM + standard magnetic resonance imaging were 80.2% and 86%, respectively. CONCLUSION: IVIM and DCE perfusion imaging could serve as promising tools for predicting treatment response in CRC patients after Neo-CRT.

Correlating Ultrastructural Changes in the Invasion Area of Colorectal Cancer with CT and MRI Imaging.

The cancer invasion of the large intestine, a destructive process that begins within the mucous membrane, causes cancer cells to gradually erode specific layers of the intestinal wall. The normal tissues of the intestine are progressively replaced by a tumour mass, leading to the impairment of the large intestine's proper morphology and function. At the ultrastructural level, the disintegration of the extracellular matrix (ECM) by cancer cells triggers the activation of inflammatory cells (macrophages) and connective tissue cells (myofibroblasts) in this area. This accumulation and the functional interactions between these cells form the tumour microenvironment (TM). The constant modulation of cancer cells and cancer-associated fibroblasts (CAFs) creates a specific milieu akin to non-healing wounds, which induces colon cancer cell proliferation and promotes their survival. This review focuses on the processes occurring at the "front of cancer invasion", with a particular focus on the role of the desmoplastic reaction in neoplasm development. It then correlates the findings from the microscopic observation of the cancer's ultrastructure with the potential of modern radiological imaging, such as computer tomography (CT) and magnetic resonance imaging (MRI), which visualizes the tumour, its boundaries, and the tissue reactions in the large intestine.

Hyperspectral imaging facilitating resect-and-discard strategy through artificial intelligence-assisted diagnosis of colorectal polyps: A pilot study.

BACKGROUND AND AIMS: The resect-and-discard strategy for colorectal polyps based on accurate optical diagnosis remains challenges. Our aim was to investigate the feasibility of hyperspectral imaging (HSI) for identifying colorectal polyp properties and diagnosis of colorectal cancer in fresh tissues during colonoscopy. METHODS: 144,900 two dimensional images generated from 161 hyperspectral images of colorectal polyp tissues were prospectively obtained from patients undergoing colonoscopy. A residual neural network model was trained with transfer learning to automatically differentiate colorectal polyps, validated by histopathologic diagnosis. The diagnostic performances of the HSI-AI model and endoscopists were calculated respectively, and the auxiliary efficiency of the model was evaluated after a 2-week interval. RESULTS: Quantitative HSI revealed histological differences in colorectal polyps. The HSI-AI model showed considerable efficacy in differentiating nonneoplastic polyps, non-advanced adenomas, and advanced neoplasia in vitro, with sensitivities of 96.0%, 94.0%, and 99.0% and specificities of 99.0%, 99.0%, and 96.5%, respectively. With the assistance of the model, the median negative predictive value of neoplastic polyps increased from 50.0% to 88.2% (p = 0.013) in novices. CONCLUSION: This study demonstrated the feasibility of using HSI as a diagnostic tool to differentiate neoplastic colorectal polyps in vitro and the potential of AI-assisted diagnosis synchronized with colonoscopy. The tool may improve the diagnostic performance of novices and facilitate the application of resect-and-discard strategy to decrease the cost.

Radiomics analysis of baseline computed tomography to predict oncological outcomes in patients treated for resectable colorectal cancer liver metastasis.

OBJECTIVE: The purpose of this study was to determine and compare the performance of pre-treatment clinical risk score (CRS), radiomics models based on computed (CT), and their combination for predicting time to recurrence (TTR) and disease-specific survival (DSS) in patients with colorectal cancer liver metastases. METHODS: We retrospectively analyzed a prospectively maintained registry of 241 patients treated with systemic chemotherapy and surgery for colorectal cancer liver metastases. Radiomics features were extracted from baseline, pre-treatment, contrast-enhanced CT images. Multiple aggregation strategies were investigated for cases with multiple metastases. Radiomics signatures were derived using feature selection methods. Random survival forests (RSF) and neural network survival models (DeepSurv) based on radiomics features, alone or combined with CRS, were developed to predict TTR and DSS. Leveraging survival models predictions, classification models were trained to predict TTR within 18 months and DSS within 3 years. Classification performance was assessed with area under the receiver operating characteristic curve (AUC) on the test set. RESULTS: For TTR prediction, the concordance index (95% confidence interval) was 0.57 (0.57-0.57) for CRS, 0.61 (0.60-0.61) for RSF in combination with CRS, and 0.70 (0.68-0.73) for DeepSurv in combination with CRS. For DSS prediction, the concordance index was 0.59 (0.59-0.59) for CRS, 0.57 (0.56-0.57) for RSF in combination with CRS, and 0.60 (0.58-0.61) for DeepSurv in combination with CRS. For TTR classification, the AUC was 0.33 (0.33-0.33) for CRS, 0.77 (0.75-0.78) for radiomics signature alone, and 0.58 (0.57-0.59) for DeepSurv score alone. For DSS classification, the AUC was 0.61 (0.61-0.61) for CRS, 0.57 (0.56-0.57) for radiomics signature, and 0.75 (0.74-0.76) for DeepSurv score alone. CONCLUSION: Radiomics-based survival models outperformed CRS for TTR prediction. More accurate, noninvasive, and early prediction of patient outcome may help reduce exposure to ineffective yet toxic chemotherapy or high-risk major hepatectomies.

Feasibility of indocyanine green (ICG) fluorescence in ex vivo pathological dissection of colorectal lymph nodes-a pilot study.

Accurate lymph node (LN) retrieval during colorectal carcinoma resection is pivotal for precise N-staging and the determination of adjuvant therapy. Current guidelines recommend the examination of at least 12 mesocolic or mesorectal lymph nodes for accurate staging. Traditional histological processing techniques, reliant on visual inspection and palpation, are time-consuming and heavily dependent on the examiner's expertise and availability. Various methods have been documented to enhance LN retrieval from colorectal specimens, including intra-arterial ex vivo methylene blue injection. Recent studies have explored the utility of indocyanine green (ICG) fluorescence imaging for visualizing pericolic lymph nodes and identifying sentinel lymph nodes in colorectal malignancies. This study included 10 patients who underwent colon resection for malignant tumors. During surgery, intravenous ICG dye and an endoscopic camera were employed to assess intestinal perfusion. Post-resection, ex vivo intra-arterial administration of ICG dye was performed on the specimens, followed by routine histological processing and an ICG-assisted lymph node dissection. The objective was to evaluate whether ICG imaging could identify additional lymph nodes compared to routine manual dissection and to assess the clinical relevance of these findings. For each patient, a minimum of 12 lymph nodes (median = 25.5, interquartile range = 12.25, maximum = 33) were examined. ICG imaging facilitated the detection of a median of three additional lymph nodes not identified during routine processing. Metastatic lymph nodes were found in four patients however no additional metastatic nodes were detected with ICG assistance. Our findings suggest that ex vivo intra-arterial administration of indocyanine green dye can augment lymph node dissection, particularly in cases where the number of lymph nodes retrieved is below the recommended threshold of 12.

Validation of an automated segmentation method for body composition analysis in colorectal cancer patients using diagnostic abdominal computed tomography images.

BACKGROUND & AIMS: Several automated programs have been developed to facilitate body composition analysis of images from abdominal computed tomography (CT) scans. External validation in patients with colorectal cancer is necessary for use in research and clinical practice. Our aim was to validate an automatic method (AutoMATiCA) of segmenting CT images at the third lumbar level (L3) from patients with colorectal cancer, by comparing with manual segmentation. METHODS: Diagnostic abdominal CT scans of consecutive patients with stage I-III colorectal cancer were analysed to measure cross-sectional areas and tissue densities of skeletal muscle and intra-muscular, visceral, and subcutaneous adipose tissue. Trained analysts performed manual segmentation of L3 CT images using SliceOmatic. Automatic segmentation was performed using AutoMATiCA, an open-source software. The Dice similarity coefficient (DSC) was calculated to assess segmentation accuracy. Agreement of automatic with manual segmentation was evaluated using intra-class correlation coefficients (ICCs) and Bland-Altman plots with limits of agreement. RESULTS: A total of 292 scans were included, of which 62% were from male patients. The agreement of AutoMATiCA with the manual segmentation was excellent, with median DSC values ranging from 0.900 to 0.991 and ICCs above 0.95 for all segmented areas. No systematic deviations were observed in Bland-Altman plots for all segmented areas, with overall narrow limits of agreement. CONCLUSIONS: AutoMATiCA provides an accurate segmentation of abdominal CT images from patients with colorectal cancer. Our findings support its use as a highly efficient automated tool for body composition analysis in research and potentially also in clinical practice.

Quantitative analysis of peri-intestinal lymph node metastasis using indocyanine green fluorescence imaging technology.

We evaluated the efficacy of indocyanine green fluorescence imaging compared to that of traditional nanocarbon dyes in assessing peri-intestinal lymph node metastasis in patients with colorectal cancer, which is a key prognostic factor. The relationship between indocyanine green fluorescence imaging and histopathological outcomes in patients with colon cancer has also been explored. A retrospective analysis was conducted on 30 patients with colon cancer (from May to October 2023) confirmed by surgical pathology. Tumors were marked with indocyanine green (ICG) or nanocarbon via colonoscopy 16 to 24 hours before surgery. Within 15 minutes after surgery, peri-intestinal lymph node fluorescence imaging and hematoxylin and eosin staining were used to assess the distribution of cancer foci. The correlation between cancer foci distribution, fluorescence intensity, and area under the receiver operating characteristic curve was measured. Among 243 metastatic lymph nodes from 30 patients, 18 were found. After the patients were divided into metastatic and nonmetastatic groups, significant differences in tumor differentiation and stage were noted (P < .001). The fluorescence intensity was strongly correlated with the presence and proportion of metastasis (area under the receiver operating characteristic curve = 0.931), whereas nanocarbon staining showed no significant correlation (P = .81). All P values were two-sided, with P < .05 indicating statistical significance. Lymph nodes with malignant intestinal tumor metastasis displayed weaker ICG fluorescence than did nonmetastatic nodes. Combining ICG and nanocarbon staining techniques enhances intraoperative lymph node dissection and postoperative analysis, indicating their potential utility in colorectal cancer surgery.

Identification of A0 minimum ablative margins for colorectal liver metastases: multicentre, retrospective study using deformable CT registration and artificial intelligence-based autosegmentation.

BACKGROUND: Several ablation confirmation software methods for minimum ablative margin assessment have recently been developed to improve local outcomes for patients undergoing thermal ablation of colorectal liver metastases. Previous assessments were limited to single institutions mostly at the place of development. The aim of this study was to validate the previously identified 5 mm minimum ablative margin (A0) using autosegmentation and biomechanical deformable image registration in a multi-institutional setting. METHODS: This was a multicentre, retrospective study including patients with colorectal liver metastases undergoing CT- or ultrasound-guided microwave or radiofrequency ablation during 2009-2022, reporting 3-year local disease progression (residual unablated tumour or local tumour progression) rates by minimum ablative margin across all institutions and identifying an intraprocedural contrast-enhanced CT-based minimum ablative margin associated with a 3-year local disease progression rate of less than 1%. RESULTS: A total of 400 ablated colorectal liver metastases (median diameter of 1.5 cm) in 243 patients (145 men; median age of 62 [interquartile range 54-70] years) were evaluated, with a median follow-up of 26 (interquartile range 17-40) months. A total of 119 (48.9%) patients with 186 (46.5%) colorectal liver metastases were from international institutions B, C, and D that were not involved in the software development. Three-year local disease progression rates for 0 mm, >0 and <5 mm, and 5 mm or larger minimum ablative margins were 79%, 15%, and 0% respectively for institution A (where the software was developed) and 34%, 19%, and 2% respectively for institutions B, C, and D combined. Local disease progression risk decreased to less than 1% with an intraprocedurally confirmed minimum ablative margin greater than 4.6 mm. CONCLUSION: A minimum ablative margin of 5 mm or larger demonstrates optimal local oncological outcomes. It is proposed that an intraprocedural minimum ablative margin of 5 mm or larger, confirmed using biomechanical deformable image registration, serves as the A0 for colorectal liver metastasis thermal ablation.

Colorectal polyp segmentation with denoising diffusion probabilistic models.

Early detection of polyps is essential to decrease colorectal cancer(CRC) incidence. Therefore, developing an efficient and accurate polyp segmentation technique is crucial for clinical CRC prevention. In this paper, we propose an end-to-end training approach for polyp segmentation that employs diffusion model. The images are considered as priors, and the segmentation is formulated as a mask generation process. In the sampling process, multiple predictions are generated for each input image using the trained model, and significant performance enhancements are achieved through the use of majority vote strategy. Four public datasets and one in-house dataset are used to train and test the model performance. The proposed method achieves mDice scores of 0.934 and 0.967 for datasets Kvasir-SEG and CVC-ClinicDB respectively. Furthermore, one cross-validation is applied to test the generalization of the proposed model, and the proposed methods outperformed previous state-of-the-art(SOTA) models to the best of our knowledge. The proposed method also significantly improves the segmentation accuracy and has strong generalization capability.

Non-invasive CT radiomic biomarkers predict microsatellite stability status in colorectal cancer: a multicenter validation study.

BACKGROUND: Microsatellite instability (MSI) status is a strong predictor of response to immunotherapy of colorectal cancer. Radiogenomic approaches promise the ability to gain insight into the underlying tumor biology using non-invasive routine clinical images. This study investigates the association between tumor morphology and the status of MSI versus microsatellite stability (MSS), validating a novel radiomic signature on an external multicenter cohort. METHODS: Preoperative computed tomography scans with matched MSI status were retrospectively collected for 243 colorectal cancer patients from three hospitals: Seoul National University Hospital (SNUH); Netherlands Cancer Institute (NKI); and Fondazione IRCCS Istituto Nazionale dei Tumori, Milan Italy (INT). Radiologists delineated primary tumors in each scan, from which radiomic features were extracted. Machine learning models trained on SNUH data to identify MSI tumors underwent external validation using NKI and INT images. Performances were compared in terms of area under the receiving operating curve (AUROC). RESULTS: We identified a radiomic signature comprising seven radiomic features that were predictive of tumors with MSS or MSI (AUROC 0.69, 95% confidence interval [CI] 0.54-0.84, p = 0.018). Integrating radiomic and clinical data into an algorithm improved predictive performance to an AUROC of 0.78 (95% CI 0.60-0.91, p = 0.002) and enhanced the reliability of the predictions. CONCLUSION: Differences in the radiomic morphological phenotype between tumors MSS or MSI could be detected using radiogenomic approaches. Future research involving large-scale multicenter prospective studies that combine various diagnostic data is necessary to refine and validate more robust, potentially tumor-agnostic MSI radiogenomic models. RELEVANCE STATEMENT: Noninvasive radiomic signatures derived from computed tomography scans can predict MSI in colorectal cancer, potentially augmenting traditional biopsy-based methods and enhancing personalized treatment strategies. KEY POINTS: Noninvasive CT-based radiomics predicted MSI in colorectal cancer, enhancing stratification. A seven-feature radiomic signature differentiated tumors with MSI from those with MSS in multicenter cohorts. Integrating radiomic and clinical data improved the algorithm's predictive performance.

Evolving trends in CT colonography: A 10-year analysis of use and associated factors.

PURPOSE: Computed tomographic colonography (CTC) is a non-invasive screening test for colorectal cancer (CRC) with high sensitivity and low risk of complications. We used a nationally representative sample of screening-eligible adults to examine trends in and factors associated with CTC use. METHODS: We examined CTC use among 58,058 adults in the National Health Interview Survey in 2010, 2015, 2018, 2019, and 2021. For each survey year, we estimated CTC use by sociodemographic and health factors. We used multivariable logistic regression to identify factors associated with CTC use. RESULTS: A total of 1.7 % adults reported receiving CTC across all survey years. CTC use was similar in 2010 (1.3 %), 2015 (0.8 %), 2018 (1.4 %), and 2019 (1.4 %) but increased in 2021 (3.5 %, p < 0.05). In multivariable analysis, survey year 2021 [vs. 2010, odds ratio (OR) 2.51, 95 % confidence interval (CI) 1.83-3.43], Hispanic (OR 1.73, 95 % CI 1.34-2.23), non-Hispanic Black (OR 2.07, 95 % CI 1.67-2.57), and household income <200 % federal poverty level (vs. >400 %, OR 1.25, 95 % CI 1.01-1.57) was associated with CTC use. Further, adults with a history of diabetes (OR 1.20, 95 % CI 1.01-1.45), chronic obstructive pulmonary disease (OR 1.58, 95 % CI 1.25-1.99), cancer (OR 1.29, 95 % CI 1.05-1.58), or past-year hospital admissions (OR 1.44, 95 % CI 1.18-1.78) were more likely to receive CTC. CONCLUSION: CTC use remained low from 2010 to 2019 but increased in 2021. CTC use was more frequent among adults with chronic health conditions, minorities, and adults with lower income, and may help reduce disparities in CRC screening.

Resting-State Brain Function Alteration in Colorectal Cancer Patients.

BACKGROUND AND PURPOSE: To investigate the abnormal pattern of altered functional activity in the brain and the neuroimaging mechanisms underlying the cognitive impairment of patients with colorectal cancer (CRC) via resting-state functional magnetic resonance imaging (rs-fMRI). MATERIALS AND METHODS: CRC patients (n = 56) and healthy controls (HCs) (n = 50) were studied. The participants underwent rs-fMRI scans and the Montreal Cognitive Assessment (MoCA). The amplitude of low-frequency fluctuations (ALFF), degree centrality (DC), regional homogeneity (ReHo), and MoCA scores, were calculated for participants. RESULTS: The scores of executives, visuospatial, memory, language and attention were lower in CRC patients. ReHo and ALFF values in the left postcentral gyrus, ReHo values in the right postcentral gyrus, ALFF and DC values in the left middle occipital gyrus, ReHo and DC values in the right lingual gyrus, DC values in the right angular gyrus and precuneus, and ALFF values in the left middle temporal gyrus decreased conspicuously in the CRC patients. CONCLUSION: CRC patients have abnormal resting state function, mainly in the brain areas involved in cognitive function. The overlapping brain regions with abnormal functional indicators are in the middle occipital gyrus, postcentral gyrus, and lingual gyrus. This study reveals the potential biological pathways involved in brain impairment and neurocognitive decline in patients with CRC.

Evaluation of the Relationship Between Portal Vein Diameter and Colorectal Liver Metastases on Computed Tomography.

BACKGROUND/AIM: The most common and often first metastatic site of colorectal cancer (CRC) is the liver, and radiological modalities have a critical role in the diagnosis of colorectal liver metastasis (CRLM). In this study, the possible relationship between portal vein diameter, number of metastases, and metastasis diameter was evaluated in CRLM patients who underwent computed tomography (CT) examination with intravenous contrast (IV). PATIENTS AND METHODS: Cases diagnosed with CRLM who underwent abdominal CT examination with IV contrast between December 2020 and January 2024 were retrospectively scanned. People over the age of 18 were included, and cases were divided into three subgroups according to the number of metastases: a (single), b (two), and c (three and/or more). RESULTS: There were 101 male and 74 female cases; the youngest case was 39 (male) and the oldest case was 87 (male) years old. According to the number of CRLMs, group a had 47 cases, group b had 23, and group c had 105 cases. The minimum diameter of metastasis was 0.74 cm, the maximum was 11.86 cm, and the mean diameter was 4.45±2.67 cm. There was a significant correlation between the presence of metastasis in the left lobe and the diameter of the metastases (p<0.05). CONCLUSION: The relationship between portal vein diameter and CRLM using contrast-enhanced CT scans was explored. While no significant correlation was found between portal vein diameters and metastasis size, a notable association was observed between metastasis size and their presence in the left liver lobe. These findings suggest that CRLMs in the left lobe may respond better to preoperative chemotherapy and surgical interventions. This novel insight could help develop targeted treatment strategies for CRLM, though further research with larger cohorts is needed.

Know your orientation: A viewpoint-aware framework for polyp segmentation.

Automatic polyp segmentation in endoscopic images is critical for the early diagnosis of colorectal cancer. Despite the availability of powerful segmentation models, two challenges still impede the accuracy of polyp segmentation algorithms. Firstly, during a colonoscopy, physicians frequently adjust the orientation of the colonoscope tip to capture underlying lesions, resulting in viewpoint changes in the colonoscopy images. These variations increase the diversity of polyp visual appearance, posing a challenge for learning robust polyp features. Secondly, polyps often exhibit properties similar to the surrounding tissues, leading to indistinct polyp boundaries. To address these problems, we propose a viewpoint-aware framework named VANet for precise polyp segmentation. In VANet, polyps are emphasized as a discriminative feature and thus can be localized by class activation maps in a viewpoint classification process. With these polyp locations, we design a viewpoint-aware Transformer (VAFormer) to alleviate the erosion of attention by the surrounding tissues, thereby inducing better polyp representations. Additionally, to enhance the polyp boundary perception of the network, we develop a boundary-aware Transformer (BAFormer) to encourage self-attention towards uncertain regions. As a consequence, the combination of the two modules is capable of calibrating predictions and significantly improving polyp segmentation performance. Extensive experiments on seven public datasets across six metrics demonstrate the state-of-the-art results of our method, and VANet can handle colonoscopy images in real-world scenarios effectively. The source code is available at https://github.com/1024803482/Viewpoint-Aware-Network.

PathoDuet: Foundation models for pathological slide analysis of H&E and IHC stains.

Large amounts of digitized histopathological data display a promising future for developing pathological foundation models via self-supervised learning methods. Foundation models pretrained with these methods serve as a good basis for downstream tasks. However, the gap between natural and histopathological images hinders the direct application of existing methods. In this work, we present PathoDuet, a series of pretrained models on histopathological images, and a new self-supervised learning framework in histopathology. The framework is featured by a newly-introduced pretext token and later task raisers to explicitly utilize certain relations between images, like multiple magnifications and multiple stains. Based on this, two pretext tasks, cross-scale positioning and cross-stain transferring, are designed to pretrain the model on Hematoxylin and Eosin (H&E) images and transfer the model to immunohistochemistry (IHC) images, respectively. To validate the efficacy of our models, we evaluate the performance over a wide variety of downstream tasks, including patch-level colorectal cancer subtyping and whole slide image (WSI)-level classification in H&E field, together with expression level prediction of IHC marker, tumor identification and slide-level qualitative analysis in IHC field. The experimental results show the superiority of our models over most tasks and the efficacy of proposed pretext tasks. The codes and models are available at https://github.com/openmedlab/PathoDuet.

Preoperative imaging of colorectal liver metastases: what the radiologist and the multidisciplinary team need to know.

The management of patients with colorectal liver metastases (CRLM) has transformed over the past 2 decades. Advances in surgical techniques, systemic therapies, and local treatments have resulted in a paradigm shift. Disease that would once have been considered terminal is now frequently treated aggressively with both a disease-free and overall survival benefit. In line with the expanding range of treatment options, there has been an increase in the volume and complexity of imaging required in the management of these patients to ensure optimal patient selection and outcome. The radiologist plays a pivotal role in interpreting these studies, conveying the relevant information and informing the discussion at multidisciplinary team meetings. The purpose of this review is to provide an update for radiologists on the current surgical management of patients with CRLM highlighting specific imaging information that is required by the multidisciplinary team when assessing resectability and/or the need for additional liver-directed therapies.

Transformers for colorectal cancer segmentation in CT imaging.

PURPOSE: Most recently transformer models became the state of the art in various medical image segmentation tasks and challenges, outperforming most of the conventional deep learning approaches. Picking up on that trend, this study aims at applying various transformer models to the highly challenging task of colorectal cancer (CRC) segmentation in CT imaging and assessing how they hold up to the current state-of-the-art convolutional neural network (CNN), the nnUnet. Furthermore, we wanted to investigate the impact of the network size on the resulting accuracies, since transformer models tend to be significantly larger than conventional network architectures. METHODS: For this purpose, six different transformer models, with specific architectural advancements and network sizes were implemented alongside the aforementioned nnUnet and were applied to the CRC segmentation task of the medical segmentation decathlon. RESULTS: The best results were achieved with the Swin-UNETR, D-Former, and VT-Unet, each transformer models, with a Dice similarity coefficient (DSC) of 0.60, 0.59 and 0.59, respectively. Therefore, the current state-of-the-art CNN, the nnUnet could be outperformed by transformer architectures regarding this task. Furthermore, a comparison with the inter-observer variability (IOV) of approx. 0.64 DSC indicates almost expert-level accuracy. The comparatively low IOV emphasizes the complexity and challenge of CRC segmentation, as well as indicating limitations regarding the achievable segmentation accuracy. CONCLUSION: As a result of this study, transformer models underline their current upward trend in producing state-of-the-art results also for the challenging task of CRC segmentation. However, with ever smaller advances in total accuracies, as demonstrated in this study by the on par performances of multiple network variants, other advantages like efficiency, low computation demands, or ease of adaption to new tasks become more and more relevant.

Diagnostic performance of the Japanese Narrow-band imaging expert team classification system using dual focus magnification in real-time Vietnamese setting.

The JNET classification, combined with magnified narrowband imaging (NBI), is essential for predicting the histology of colorectal polyps and guiding personalized treatment strategies. Despite its recognized utility, the diagnostic efficacy of JNET classification using NBI with dual focus (DF) magnification requires exploration in the Vietnamese context. This study aimed to investigate the diagnostic performance of the JNET classification with the NBI-DF mode in predicting the histology of colorectal polyps in Vietnam. A cross-sectional study was conducted at the University Medical Center in Ho Chi Minh City, Vietnam. During real-time endoscopy, endoscopists evaluated the lesion characteristics and recorded optical diagnoses using the dual focus mode magnification according to the JNET classification. En bloc lesion resection (endoscopic or surgical) provided the final pathology, serving as the reference standard for optical diagnoses. A total of 739 patients with 1353 lesions were recruited between October 2021 and March 2023. The overall concordance with the JNET classification was 86.9%. Specificities and positive predictive values for JNET types were: type 1 (95.7%, 88.3%); type 2A (81.4%, 90%); type 2B (96.6%, 54.7%); and type 3 (99.9%, 93.3%). The sensitivity and negative predictive value for differentiating neoplastic from non-neoplastic lesions were 97.8% and 88.3%, respectively. However, the sensitivity for distinguishing malignant from benign neoplasia was lower at 64.1%, despite a specificity of 95.9%. Notably, the specificity and positive predictive value for identifying deep submucosal cancer were high at 99.8% and 93.3%. In Vietnam, applying the JNET classification with NBI-DF demonstrates significant value in predicting the histology of colorectal polyps. This classification guides treatment decisions and prevents unnecessary surgeries.