Advances in endoscopy for colorectal polyp detection and classification

Computer-aided colorectal cancer diagnosis: AI-driven image segmentation and classification

Colorectal cancer is an enormous health concern since it is among the most lethal types of malignancy. The manual examination has its limitations, including subjectivity and data overload. To overcome these challenges, computer-aided diagnostic systems focusing on image segmentation and abnormality classification have been developed. This study presents a two-stage approach for the automatic detection of five types of colorectal abnormalities in addition to a control group: polyp, low-grade intraepithelial neoplasia, high-grade intraepithelial neoplasia, serrated adenoma, adenocarcinoma. In the first stage, UNet3+ was used for image segmentation to locate the anomalies, while in the second stage, the Cross-Attention Multi-Scale Vision Transformer deep learning model was used to predict the type of anomaly after highlighting the anomaly on the raw images. In anomaly segmentation, UNet3+ achieved values of 0.9872, 0.9422, 0.9832, and 0.9560 for Dice Coefficient, Jaccard Index, Sensitivity, Specificity respectively. In anomaly detection, the Cross-Attention Multi-Scale Vision Transformer model attained a classification performance of 0.9340, 0.9037, 0.9446, 0.8723, 0.9102, 0.9849 for accuracy, F1 score, precision, recall, Matthews correlation coefficient, and specificity, respectively. The proposed approach proves its capacity to alleviate the overwhelm of pathologists and enhance the accuracy of colorectal cancer diagnosis by achieving high performance in both the identification of anomalies and the segmentation of regions.

Precision in detecting colon lesions: A key to effective screening policy but will it improve overall outcomes?

Colonoscopy is the gold standard for the screening and diagnosis of colorectal cancer, resulting in a decrease in the incidence and mortality of colon cancer. However, it has a 21% rate of missed polyps. Several strategies have been devised to increase polyp detection rates and improve their characterization and delimitation. These include chromoendoscopy (CE), the use of other devices such as Endo cuffs, and major advances in endoscopic equipment [high definition, magnification, narrow band imaging, i-scan, flexible spectral imaging color enhancement, texture and color enhancement imaging (TXI), etc.]. In the retrospective study by Hiramatsu et al, they compared white-light imaging with CE, TXI, and CE + TXI to determine which of these strategies allows for better definition and delimitation of polyps. They concluded that employing CE associated with TXI stands out as the most effective method to utilize. It remains to be demonstrated whether these results are extrapolatable to other types of virtual CE. Additionally, further investigation is needed in order to ascertain whether this strategy could lead to a reduction in the recurrence of excised lesions and potentially lower the occurrence of interval cancer.

Spatio-Temporal Feature Transformation Based Polyp Recognition for Automatic Detection: Higher Accuracy than Novice Endoscopists in Colorectal Polyp Detection and Diagnosis

BACKGROUND

Artificial intelligence represents an emerging area with promising potential for improving colonoscopy quality.

AIMS

To develop a colon polyp detection model using STFT and evaluate its performance through a randomized sample experiment.

METHODS

Colonoscopy videos from the Digestive Endoscopy Center of the First Affiliated Hospital of Anhui Medical University, recorded between January 2018 and November 2022, were selected and divided into two datasets. To verify the model's practical application in clinical settings, 1500 colonoscopy images and 1200 polyp images of various sizes were randomly selected from the test set and compared with the STFT model's and endoscopists' recognition results with different years of experience.

RESULTS

In the randomized sample trial involving 1500 colonoscopy images, the STFT model demonstrated significantly higher accuracy and specificity compared to endoscopists with low years of experience (0.902 vs. 0.809, 0.898 vs. 0.826, respectively). Moreover, the model's sensitivity was 0.904, which was higher than that of endoscopists with low, medium, or high years of experience (0.80, 0.896, 0.895, respectively), with statistical significance (P < 0.05). In the randomized sample experiment of 1200 polyp images of different sizes, the accuracy of the STFT model was significantly higher than that of endoscopists with low years of experience when the polyp size was ≤ 0.5 cm and 0.6-1.0 cm (0.902 vs. 0.70, 0.953 vs. 0.865, respectively).

CONCLUSIONS

The STFT-based colon polyp detection model exhibits high accuracy in detecting polyps in colonoscopy videos, with a particular efficiency in detecting small polyps (≤ 0.5 cm)(0.902 vs. 0.70, P < 0.001).

Effect of add-on devices with projections on screening colonoscopy: a systematic review and meta-analysis

Background

Add-on devices with projections, e.g., Endocuff, Endocuff Vision, EndoRings, and Wingcap, placed on the distal tip of the colonoscope promise to improve the detection of precancerous lesions. We performed a meta-analysis to evaluate the performance of these devices exclusively among individuals undergoing colonoscopy for screening purpose.

Methods

A computerized literature search was performed across MEDLINE and Cochrane Library databases for randomized controlled trials that compared standard colonoscopy (SC) to procedures using add-on devices. The primary outcome was adenoma detection rate (ADR), while secondary outcomes included polyp detection rate (PDR), advanced ADR (AADR), and sessile serrated lesion detection rate (SSLDR). The effect size on study outcomes was calculated using a random-effects model and presented as the risk ratio (RR) and 95% confidence interval (CI).

Results

Seven studies enrolling a total of 5785 patients were included. The use of add-on-devices with projections was associated with a higher ADR compared to SC: 45.9% vs. 41.1%; RR 1.18, 95%CI 1.02-1.37; P=0.03; I2=79%. Although PDR was higher in screening colonoscopies assisted by add-on devices as compared to SC, the difference failed to reach significance: 55.1% vs. 50.8%; RR 1.10, 95%CI 0.96-1.26; P=0.17; I2=75%. No difference was found between procedures assisted by add-on devices with projections and SC colonoscopies in terms of AADR (18.5% vs. 17.6%; RR 1.00, 95%CI 0.79-1.27; P=0.98; I2=56%) or SSLDR (6.8% vs. 5.8%; RR 1.17, 95%CI 0.95-1.44; P=0.15; I2=0%).

Conclusion

Colonoscopy assisted by add-on devices with projections achieves a better ADR compared to SC among individuals undergoing screening for bowel cancer.

EBHI-Seg: A novel enteroscope biopsy histopathological hematoxylin and eosin image dataset for image segmentation tasks

Background and purpose

Colorectal cancer is a common fatal malignancy, the fourth most common cancer in men, and the third most common cancer in women worldwide. Timely detection of cancer in its early stages is essential for treating the disease. Currently, there is a lack of datasets for histopathological image segmentation of colorectal cancer, which often hampers the assessment accuracy when computer technology is used to aid in diagnosis.

Methods

This present study provided a new publicly available Enteroscope Biopsy Histopathological Hematoxylin and Eosin Image Dataset for Image Segmentation Tasks (EBHI-Seg). To demonstrate the validity and extensiveness of EBHI-Seg, the experimental results for EBHI-Seg are evaluated using classical machine learning methods and deep learning methods.

Results

The experimental results showed that deep learning methods had a better image segmentation performance when utilizing EBHI-Seg. The maximum accuracy of the Dice evaluation metric for the classical machine learning method is 0.948, while the Dice evaluation metric for the deep learning method is 0.965.

Conclusion

This publicly available dataset contained 4,456 images of six types of tumor differentiation stages and the corresponding ground truth images. The dataset can provide researchers with new segmentation algorithms for medical diagnosis of colorectal cancer, which can be used in the clinical setting to help doctors and patients. EBHI-Seg is publicly available at: https://figshare.com/articles/dataset/EBHI-SEG/21540159/1.

Induction of apoptosis and cell cycle arrest in colorectal cancer cells by novel anticancer metabolites of Streptomyces sp. 801

Background

Colorectal cancer is the third and most significant cause of death and fourth most common cancer in the world. Chemotherapy can be introduced in the cases of locally or distantly invasive colorectal cancer. In recent years Actinomycetes, especially the genus Streptomyces, contain numerous bioactive compounds, some of which are known as important anti-tumor chemotherapy drugs. In this research, we aimed to explore the anti-cancer mode of action of Streptomyces sp. 801 on colorectal cancer cells in vitro conditions.

Methods

Fermented supernatant of strain Streptomyces sp. 801 isolated from soil showed maximum growth inhibition on human colorectal cancer cells. The cytotoxic effects of various concentrations of EtOAc extract from bacterial culture supernatant on HT-29, HCT 116 and SW480 cancer cells were surveyed using the MTT assay. Moreover, flow cytometry assays and Bax, Bcl-2, Cyclin D1 and P21 gene expressions were carried out to assess the apoptotic and cell cycle effects. Also, the scratch assay was performed to measure migration. Finally, Ethyl acetate (EtOAc) extract was analyzed by LC–MS to identify anti-cancer compounds.

Results

The cell viability of all three cell lines were decreased in a dose-dependent manner. The successful induction of apoptosis and cell cycle arrest at IC₅₀ values, were confirmed by flow cytometry as well as by the mRNA expression levels of the genes involved in these processes. Scratch assays indicated the inhibition of cell migration in the cancer cell lines treated by Streptomyces sp. 801. Nine anti-cancer compounds of Streptomyces sp. 801 were detected by liquid chromatography–mass spectrometry (LC–MS) analysis.

Conclusions

These findings suggest that Streptomyces sp. 801 can be a source of promising anticancer metabolites.

Graphical Abstract

Targeted Nursing Combined with Endoscopic Submucosal Injection of Carbon Nanoparticles in the Treatment of Colorectal Cancer

Aim

To evaluate the effectiveness of targeted nursing in an endoscopic submucosal injection of carbon nanoparticles to locate colorectal cancer.

Methods

From September 2017 to September 2019, 82 patients with colorectal cancer who underwent endoscopic submucosal injection of carbon nanoparticles for locating the tumor were recruited and assigned via the random number table method (1 : 1) to receive either conventional nursing (control group) or targeted nursing (observation group). Outcome measures included psychological status, compliance, nursing satisfaction, quality of life, and daily living ability.

Results

After intervention, the self-rating anxiety scale (SAS) scores and self-rating depression scale (SDS) scores were decreased in both groups, with lower results in the observation group (P < 0.001). Patients given target nursing were associated with higher compliance and nursing satisfaction of patients versus conventional nursing (P < 0.05). Patients receiving targeted nursing had a better quality of life versus those receiving conventional nursing (P < 0.001). Targeted nursing resulted in a higher Barthel index (BI) in patients versus conventional nursing 1 d, 7 d, and 14 d after nursing (P < 0.05).

Conclusion

Targeted nursing alleviates the negative emotions of patients with colorectal cancer and improves their compliance, nursing satisfaction, daily living ability, and quality of life.

Bioactive Components From Gracilaria rubra With Growth Inhibition on HCT116 Colon Cancer Cells and Anti-inflammatory Capacity in RAW 264.7 Macrophages

Gracilaria rubra is rich in bioactive compounds with various potential health benefits. This study aimed to elucidate the profile of both extractable bioactive components (EBCs) and non-extractable bioactive components (NEBCs) of G. rubra and determine their anti-colon cancer and anti-inflammatory activities. Both EBCs and NEBCs displayed strong suppressive effects on the viability of HCT116 cells, which causes cell cycle arrest, induces cellular apoptosis, and regulates the expression of cyclin-dependent kinases (CDKs) and tumor suppressor proteins. Additionally, EBCs and NEBCs from G. rubra displayed anti-inflammatory functions via inhibiting the production of nitric oxide (NO), reactive oxygen species (ROS), and proinflammatory cytokines in activated macrophages and regulating the expression levels of cyclooxygenase-2 (COX-2), inducible nitric oxide synthase (iNOS), NADPH-quinone oxidoreductase-1 (NQO-1), and heme oxygenase 1 (HO-1). These findings provide a rationale for animal and human studies designed to evaluate the chemopreventive and anti-inflammatory potential of these bioactive compounds from G. rubra.

Evaluation of an image enhancement system for the assessment of nasal and paranasal sinus diseases

PURPOSE

Dysplasia and cancer of the upper aerodigestive tract are characterized by significant neoangiogenesis. This can be recognized by optical methods like the Storz Professional Image Enhancement System (SPIES). Up to now, there are no reports of using this novel technique for examining nasal diseases. The objective of this study was to evaluate the use of SPIES during sinus surgery to help differentiate various nasal pathologies and determine their extension.

METHODS

Patients (n = 27) with different pathologies in the region of the paranasal sinuses were operated via functional endoscopic surgery using a 2D-HD-camera with white light and SPIES. In addition, 10 healthy individuals were examined. The system was evaluated using two different questionnaires.

RESULTS

The handling and operation of SPIES was intuitive and easy. Use of SPIES did not prolong the procedure. There was no disturbing image distortion. SPIES seemed to improve the visualization, differentiation and evaluation of vascularization of paranasal pathologies and allowed for precise and accurate surgery. Compared to examination with the 2D-HD-camera and white light alone, SPIES appeared to facilitate the identification of mucosal pathologies.

CONCLUSION

SPIES could be a promising adjunct tool to evaluate nasal pathologies intraoperatively. Especially in the case of vascularized tumors the enhanced image endoscopy seemed to be clearly superior to standard white light alone. In our study, the system facilitated the assessment of tumor extension and vascularization as well as the differentiation of healthy mucosa. Future randomized studies will be necessary to prove the potential of integrating this novel technique into the clinical routine for the differentiation of nasal pathologies and the improvement of resection margins during nasal tumor surgery.

Novel biliopancreatic duct endoscope combining optical coherence tomography with intraductal US for exploring the bile duct: a diagnostic study in a porcine model

BACKGROUND AND AIMS

Existing biliopancreatic duct endoscopy is deficient in the examination of early biliary and pancreatic tumors. This study aimed to evaluate the usefulness and feasibility of a novel ultrafine, separable, biliopancreatic duct endoscopy device with dual modalities of intraductal US (IDUS) and optical coherence tomography (OCT) for the in vivo assessment of the biliopancreatic duct system during ERCP.

METHODS

Five Bama miniature pigs were selected to probe their common bile duct and branches by using this novel equipment during ERCP. The feasibility of the procedure was evaluated by clear, clinically interpretable images obtained by using spiral scanning with the pull-back method. The clinical usefulness of the novel product was evaluated by postoperative choledochoscopy and assessment of the animal's general condition.

RESULTS

One hundred forty-one pairs of images from 5 Bama miniature pigs were acquired. Visualization of the bile duct using both OCT and IDUS was characterized by a differentiated 3-layer architecture, whereas IDUS had poor clarity when compared with OCT. Postoperative choledochoscopy showed no local lesion in the bile duct wall, and the general condition of animals was normal.

CONCLUSIONS

This prospective evaluation indicated the feasibility and potential clinical value of the novel, ultrafine, separable biliopancreatic duct endoscopy device. The fusion of the 2 imaging modalities can shed light on the early diagnosis of biliary and pancreatic tumors. Further studies will be carried out to establish diagnosis criteria with the dual-modality imaging using an animal pathologic model and a human clinical study.

Analysis of dynamic molecular networks: the progression from colorectal adenoma to cancer

BACKGROUND

Colorectal cancer (CRC) is one of the deadliest cancers worldwide. It is the fourth most deadly cancer in the world with nearly 900,000 people die every year, the progression of polyps into cancer as one of its most common developmental pathways.

METHODS

This study obtained gene chip data collections from the Gene Expression Omnibus for colorectal adenoma (GSE8671) and colorectal cancer (GSE32323). Differentially expressed genes (DEGs) in normal tissue and different stages of CRC were analyzed for clustering, comparison, and visualization using R software. The Cytoscape plugin DyNetViewer was used to construct a dynamic protein-protein interaction network. Subsequently, through the Database for Annotation, Visualization and Integrated Discovery, the DEGs were functionally annotated and path enriched.

RESULTS

Our study found that the matrix metalloprotein family and chemokines were the key regulatory genes that drove CRC progression. The Wnt signaling pathway, chemokine signaling pathway, and CRC pathway were the pathological pathways for CRC. Maintenance played an important role in this process. In addition, the related nodes and pathways at various stages may be potential mechanisms for promoting dynamic CRC progression.

CONCLUSIONS

Our study provides a better understanding of the dynamic pattern of molecular interaction networks during CRC progression and provides relevant markers for more accurate screening of cancer in polyps.

Multidisciplinary management of early rectal cancer - The role of surgical local excision in current and future clinical practice

The implementation of bowel cancer screening programs has led to a rise in the incidence of early rectal cancer. The combination of increased incidence and the growing interest in organ-sparing treatment options has led to an amplified importance of local excision techniques in treatment strategies for early rectal cancer. In addition, developments in new technologies of single-port surgery have popularized surgical techniques. Although local treatment of early rectal cancer seems promising, a multidisciplinary approach is necessary and awareness of the oncological robustness is warranted to enable shared decision-making. This review illustrates the position of surgical local excision in the treatment of early rectal cancer and reflects on its role in current and future clinical practice.

Oral indigo carmine for the detection of colon adenoma

OBJECTIVES

Colonoscopy is currently considered the optimal method to detect colorectal neoplasia; however, some adenomas remain undetected. While indigo carmine staining with a dye-spray catheter has demonstrated promising results for reducing the miss rate, we investigated the oral indigo carmine method. The aim of this study was to determine whether oral indigo carmine intake before standard colonoscopy increases the adenoma (and adenocarcinoma) detection rate (ADR) or the mean number of adenomas per patient (MAP).

MATERIALS AND METHODS

The oral indigo carmine method was performed from April 2018 to July 2020 in two hospitals. Data were collected in a prospective manner and compared to the conventional group whose data were collected retrospectively and consecutively from January 2016 to March 2018. All data were anonymized.

RESULTS

Among the 704 patients included, colonoscopies were completely performed in 693 patients (347 in the indigo group). The ADR did not significantly differ between the groups: 42.3% vs. 40.3% (indigo vs. conventional group; odds ratio: 1.13; 95% confidence interval: 0.9-1.33, p = .33). The MAP was significantly greater in the indigo group (1.15) than that in the conventional group (0.82; p = .009). The cecal intubation rate and time to cecal intubation did not differ between the indigo and conventional group (98.6% vs. 98.3%, p = .83, and 6.2 min vs. 5.9 min, p = .39, respectively).

CONCLUSION

The routine use of oral indigo carmine does not lead to a higher ADR despite the higher MAP.

Noninvasive strategies for colorectal cancer screening: opportunities and limitations

PURPOSE OF REVIEW

In the United States, only 67% of patients are up to date with colorectal cancer (CRC) screening. While colonoscopy is highly sensitive and specific for CRC and precursor lesion detection and removal, it is invasive, expensive and resource heavy. Hence, there is an unfulfilled need for multiple modality CRC screening that can improve current CRC screening rates and may be resource effective strategies when used in conjunction with a colonoscopy program. Our review highlights the complementary, often underutilized, noninvasive CRC screening methods with a focus on performance, risks, benefits, and recent updates.

RECENT FINDINGS

Studies demonstrate that fecal immunochemical testing (FIT) is superior to guaiac-based fecal occult blood tests for CRC screening. Studies show superiority of multitarget stool DNA test to FIT in sensitivity, though with concern for decreased specificity in setting of one-time tests. Technical advances continue to improve accuracy of colon capsule endoscopy. There are ongoing studies to characterize often difficult-to-detect high-risk lesions in computed tomography colonography. Septin 9 continues to have suboptimal accuracy for CRC screening, but has been shown to be associated with more advanced, invasive CRC stages.

SUMMARY

There are ongoing advances in noninvasive screening modalities for CRC; these should be considered as alternatives to colonoscopy in specific patient populations.