Integrating artificial intelligence techniques for advancements in colorectal cancer management: navigating past and predicting future direction.

Artificial Intelligence (AI) in the last few years has emerged as a valuable tool in managing colorectal cancer, revolutionizing its management at different stages. In early detection and diagnosis, AI leverages its prowess in imaging analysis, scrutinizing CT scans, MRI, and colonoscopy views to identify polyps and tumors. This ability enables timely and accurate diagnoses, initiating treatment at earlier stages. AI has helped in personalized treatment planning because of its ability to integrate diverse patient data, including tumor characteristics, medical history, and genetic information. Integrating AI into clinical decision support systems guarantees evidence-based treatment strategy suggestions in multidisciplinary clinical settings, thus improving patient outcomes. This narrative review explores the multifaceted role of AI, spanning early detection of colorectal cancer, personalized treatment planning, polyp detection, lymph node evaluation, cancer staging, robotic colorectal surgery, and training of colorectal surgeons.

Establishment and validation of an artificial intelligence-based model for real-time detection and classification of colorectal adenoma.

Colorectal cancer (CRC) prevention requires early detection and removal of adenomas. We aimed to develop a computational model for real-time detection and classification of colorectal adenoma. Computationally constrained background based on real-time detection, we propose an improved adaptive lightweight ensemble model for real-time detection and classification of adenomas and other polyps. Firstly, we devised an adaptive lightweight network modification and effective training strategy to diminish the computational requirements for real-time detection. Secondly, by integrating the adaptive lightweight YOLOv4 with the single shot multibox detector network, we established the adaptive small object detection ensemble (ASODE) model, which enhances the precision of detecting target polyps without significantly increasing the model's memory footprint. We conducted simulated training using clinical colonoscopy images and videos to validate the method's performance, extracting features from 1148 polyps and employing a confidence threshold of 0.5 to filter out low-confidence sample predictions. Finally, compared to state-of-the-art models, our ASODE model demonstrated superior performance. In the test set, the sensitivity of images and videos reached 87.96% and 92.31%, respectively. Additionally, the ASODE model achieved an accuracy of 92.70% for adenoma detection with a false positive rate of 8.18%. Training results indicate the effectiveness of our method in classifying small polyps. Our model exhibits remarkable performance in real-time detection of colorectal adenomas, serving as a reliable tool for assisting endoscopists.

Association between Colonoscopy Sedation Type and Polyp Detection: A Registry-based Cohort Study.

BACKGROUND: Colorectal cancer is a leading cause of cancer-related death. Adenomas and serrated polyps are precursors of colorectal cancer, with serrated polyps being more difficult to detect during colonoscopy. The relationship between propofol use and polyp detection remains unclear. The authors investigated the association of propofol-based versus mild-moderate sedation on adenoma and serrated polyp detection during colonoscopy. METHODS: This retrospective cohort study used observational data from the New Hampshire Colonoscopy Registry. Patients aged greater than 50 yr with screening or surveillance colonoscopies between January 1, 2015, and February 28, 2020, were included. Exclusions were diagnostic examinations, no sedation, missing pathology data, and poor bowel preparation. Multivariate logistic regression was used to evaluate differences in polyp detection between propofol and moderate sedation in the full sample while adjusting for covariates. Propensity score adjustment and clustering at the endoscopist level were used in a restricted sample analysis that included endoscopists and facilities with between 5% and 95% propofol sedation use. RESULTS: A total of 54,063 colonoscopies were analyzed in the full sample and 18,998 in the restricted sample. Serrated polyp prevalence was significantly higher using propofol (9,957 of 29,312; 34.0% [95% CI, 33.4 to 34.5%]) versus moderate sedation (6,066 of 24,751; 24.5% [95% CI, 24.0 to 25.1%]) in the full sample and restricted samples (1,410 of 4,661; 30.3% [95% CI, 28.9 to 31.6%] vs. 3,690 of 14,337; 25.7% [95% CI, 25.0 to 26.5%]). In the full sample multivariate logistic regression, propofol was associated with higher neoplasm (adjusted odds ratio, 1.25 [95% CI, 1.21 to 1.29]), adenoma (odds ratio, 1.07 [95% CI, 1.03 to 1.11]), and serrated polyp detection (odds ratio, 1.51 [95% CI, 1.46 to 1.57]). In the restricted sample using inverse probability of treatment weighted propensity score adjustment and clustering at the endoscopist level, an attenuated but statistically significant effect size was observed for serrated polyps (odds ratio, 1.13 [95% CI, 1.07 to 1.19]), but not for adenomas (odds ratio, 1.00 [95% CI, 0.95 to 1.05]) or any neoplastic lesion (odds ratio, 1.03 [95% CI, 0.98 to 1.08]). CONCLUSIONS: Propofol sedation during colonoscopy may be associated with improved detection of serrated polyps, but not adenomas.

Evaluation and Management of Malignant Colorectal Polyps.

The detection rate of dysplastic colorectal polyps has significantly increased with improved screening programs. Treatment of dysplastic polyps attempt to limit morbidity of a procedure while also considering the risk of occult lymph node metastasis. Therefore, a variety of methods have been developed to predict the rate of lymph node metastasis to help identify the optimal treatment of patients. These include both the endoscopic and pathologic assessment of the lesion. In order to reduce the morbidity of surgery for patients with low-risk lesions, multiple endoscopic therapies have been developed, including endoscopic mucosal resection, endoscopic submucosal dissection, endoscopic intermuscular dissection, and transanal endoscopic surgery.

Adenoma characteristics in the English Bowel Cancer Screening Programme.

AIM: The English Bowel Cancer Screening Programme detects colorectal cancers and premalignant polyps in a faecal occult blood test-positive population. The aim of this work is to describe the detection rates and characteristics of adenomas within the programme, identify predictive factors influencing the presence or absence of carcinoma within adenomas and identify the factors predicting the presence of advanced colonic neoplasia in different colon segments. METHOD: The Bowel Cancer Screening System was retrospectively searched for polyps detected during colonoscopies between June 2006 and June 2012, at which time a guaiac test was being used. Data on size, location and histological features were collected, and described. Univariate and multivariate analyses were used to determine the significant factors influencing the development of carcinoma within an adenoma. RESULTS: A total of 229 419 polyps were identified; after exclusions 136 973 adenomas from 58 334 patients were evaluated. Over half were in the rectum or sigmoid colon. Subcentimetre adenomas accounted for 69.8% of the total. The proportion of adenomas containing advanced histological features increased with increasing adenoma size up to 35 mm, then plateaued. A focus of carcinoma was found in 2282 (1.7%) adenomas, of which 95.6% were located distally. Carcinoma was identified even in diminutive adenomas (0.1%). The proportion of adenomas containing cancer was significantly higher in women than men (2.0% vs. 1.5%, p < 0.001). CONCLUSION: This national, prospectively captured dataset adds robust information about histological features of adenomas that convey an increased risk for colorectal cancer, and identifies caecal adenomas, high-grade dysplasia, increasing adenoma size, distal location and female sex as independent risk factors associated with carcinoma.

Recurrence and Carcinogenetic Rates of Colorectal Polyps.

AIM: to determin the recurrence rate of benign recto-colonic polyps in a 5-year interval, and compare the development rate of intrapolypoid carcinomatous lesions in polypectomized versus nonpolypectomized subjects. MATERIAL AND METHOD: a group of 77 patients diagnosed with recto-colonic polypoid lesions during the period 2014-2019 underwent colonoscopy at the time of study initiation and then annually during a five-year interval. Results: The recurrence rate of polyps increased annually from 5 to 12.5%; the highest rate was noted in the last two years. The five-year cumulative risk of neoplastic lesions was 73% in patients without polypectomy and 20% among those with endoscopic resection (p 0.05). Comparing the recurrence rate of benign lesions (60%) in patients without neoplastic findings with the recurrence rate of adenomas in patients with benign lesions (40%), a higher risk of recurrence was found in the first category, and seemed to be influenced by the personal history of pre-existing adenomatous lesions. CONCLUSION: an increased risk of colorectal polyps recurrence was reported during five year follow up; moreover, during the first three years an increased risk of malignant transformation was observed among cases in which endoscopic resection was not feasible when compared to those in which complete excision was feasible.

The Association of Metabolic Risk Factors with Advanced Adenomas in Hispanic Patients.

BACKGROUND: Obesity and metabolic syndrome (MetS) have been implicated as rising risk factors for the development of colorectal cancers. A rapid increase in the prevalence of obesity and severe obesity among Hispanic patients in the United States may present substantially increased risk for advanced colorectal neoplasia in this population. Currently, there is very little research in this area. AIMS: We sought to identify metabolic risk factors for advanced adenomas (AA) in Hispanic Americans. METHODS: We retrospectively reviewed data from the Los Angeles General (LAG) Medical Center of asymptomatic Hispanic patients above 45 years of age who underwent their first colonoscopies following a positive screening FBT. Patient demographics, metabolic characteristics, as well as colon polyp size and histology were recorded. Polyps were classified as adenomas or AA (including both high-risk adenomas and high-risk serrated polyps). Relative risk for AA was assessed by multivariate logistical regression analyses. RESULTS: Of the 672 patients in our study, 41.4% were male, 67% had adenomas, and 16% had AA. The mean BMI was 31.2 kg/m2. The mean HDL-C was 49.5 mg/dL (1.28 mmol/L) and the mean triglyceride level was 151 mg/dL. 44.6% had diabetes and 64.1% had hypertension. When comparing patients with AA to patients with no adenoma, male sex, BMI > 34.9 kg/m2, and elevated fasting triglyceride levels were associated with an increased risk of AA. FIB-4 ≥1.45 was also associated with an increased risk of AA in males. There was no significant difference in the risk of AA with diabetes, hypertension, FIB-4 score, LDL-C level, and HDL-C level. CONCLUSIONS: Hispanic patients with a positive FBT were observed to have a high incidence of AA. Class II obesity (BMI ≥ 35 kg/m2), elevated triglyceride levels were identified as risk factors among males in our study. Early interventions to address these modifiable risk factors in at-risk populations, such as multi-disciplinary weight management programs for the treatment of obesity and related co-morbidities, could potentially lead to risk reduction and CRC prevention.

The frequency of colorectal lesions in the first-degree relatives of patients with colorectal lesions among PERSIAN Guilan Cohort Study population (PGCS).

BACKGROUND: This study aimed to investigate the frequency of colorectal lesions in the first-degree relatives of patients with colorectal lesions among the Prospective Epidemiological Research Studies in Iran (PERSIAN )Guilan Cohort Study (PGCS) population. METHODS: In this cross-sectional study, 162 first-degree relatives with a history of colorectal lesions were randomly selected from 52 participants in PGCS. All subjects underwent total colonoscopy by a gastroenterologist, and a pathologist evaluated colorectal biopsies. Also, individuals' demographic information, clinical data, and dietary habits were recorded. RESULTS: The mean age of the participants was 56.55 ± 7.04. Of 86 colon polyps, 52 neoplastic and 34 non-neoplastic polyps were observed in 56 patients (34.6%). Individuals with age > 60 years had 3.29-fold increased odds of developing colorectal polyps (OR = 3.29, 95% CI: 1.13-9.56, P = 0.029). The smokers were 2.73 times more susceptible to developing colorectal polyps than non-smokers (OR = 2.73, 95% CI: 1.24-6.02, P = 0.013). Moreover, consumption of vegetables more than three times per day was associated with decreased OR of colorectal polyp development (OR = 0.43, CI: 0.19-0.98, P = 0.045). CONCLUSIONS: Considering the high prevalence of neoplastic colorectal polyps among the first-degree relatives of patients with colorectal lesions, early screening is recommended for individuals with a family history of colorectal lesions.

Colon Polyp Surveillance: Separating the Wheat From the Chaff.

One goal of colorectal cancer (CRC) screening is to prevent CRC incidence by removing precancerous colonic polyps, which are detected in up to 50% of screening examinations. Yet, the lifetime risk of CRC is 3.9%-4.3%, so it is clear that most of these individuals with polyps would not develop CRC in their lifetime. It is, therefore, a challenge to determine which individuals with polyps will benefit from follow-up, and at what intervals. There is some evidence that individuals with advanced polyps, based on size and histology, benefit from intensive surveillance. However, a large proportion of individuals will have small polyps without advanced histologic features (ie, "nonadvanced"), where the benefits of surveillance are uncertain and controversial. Demand for surveillance will further increase as more polyps are detected due to increased screening uptake, recent United States recommendations to expand screening to younger individuals, and emergence of polyp detection technology. We review the current understanding and clinical implications of the natural history, biology, and outcomes associated with various categories of colon polyps based on size, histology, and number. Our aims are to highlight key knowledge gaps, specifically focusing on certain categories of polyps that may not be associated with future CRC risk, and to provide insights to inform research priorities and potential management strategies. Optimization of CRC prevention programs based on updated knowledge about the future risks associated with various colon polyps is essential to ensure cost-effective screening and surveillance, wise use of resources, and inform efforts to personalize recommendations.

Linked-color imaging with or without artificial intelligence for adenoma detection: a randomized trial.

BACKGROUND: Adenoma detection rate (ADR) is an important indicator of colonoscopy quality and colorectal cancer incidence. Both linked-color imaging (LCI) with artificial intelligence (LCA) and LCI alone increase adenoma detection during colonoscopy, although it remains unclear whether one modality is superior. This study compared ADR between LCA and LCI alone, including according to endoscopists' experience (experts and trainees) and polyp size. METHODS: Patients undergoing colonoscopy for positive fecal immunochemical tests, follow-up of colon polyps, and abdominal symptoms at a single institution were randomly assigned to the LCA or LCI group. ADR, adenoma per colonoscopy (APC), cecal intubation time, withdrawal time, number of adenomas per location, and adenoma size were compared. RESULTS: The LCA (n=400) and LCI (n=400) groups showed comparable cecal intubation and withdrawal times. The LCA group showed a significantly higher ADR (58.8% vs. 43.5%; P<0.001) and mean (95%CI) APC (1.31 [1.15 to 1.47] vs. 0.94 [0.80 to 1.07]; P<0.001), particularly in the ascending colon (0.30 [0.24 to 0.36] vs. 0.20 [0.15 to 0.25]; P=0.02). Total number of nonpolypoid-type adenomas was also significantly higher in the LCA group (0.15 [0.09 to 0.20] vs. 0.08 [0.05 to 0.10]; P=0.02). Small polyps (≤5, 6-9mm) were detected significantly more frequently in the LCA group (0.75 [0.64 to 0.86] vs. 0.48 [0.40 to 0.57], P<0.001 and 0.34 [0.26 to 0.41] vs. 0.24 [0.18 to 0.29], P=0.04, respectively). In both groups, ADR was not significantly different between experts and trainees. CONCLUSIONS: LCA was significantly superior to LCI alone in terms of ADR.

Evaluating false-positive detection in a computer-aided detection system for colonoscopy.

BACKGROUND AND AIM: Computer-aided detection (CADe) systems can efficiently detect polyps during colonoscopy. However, false-positive (FP) activation is a major limitation of CADe. We aimed to compare the rate and causes of FP using CADe before and after an update designed to reduce FP. METHODS: We analyzed CADe-assisted colonoscopy videos recorded between July 2022 and October 2022. The number and causes of FPs and excessive time spent by the endoscopist on FP (ET) were compared pre- and post-update using 1:1 propensity score matching. RESULTS: During the study period, 191 colonoscopy videos (94 and 97 in the pre- and post-update groups, respectively) were recorded. Propensity score matching resulted in 146 videos (73 in each group). The mean number of FPs and median ET per colonoscopy were significantly lower in the post-update group than those in the pre-update group (4.2 ± 3.7 vs 18.1 ± 11.1; P < 0.001 and 0 vs 16 s; P < 0.001, respectively). Mucosal tags, bubbles, and folds had the strongest association with decreased FP post-update (pre-update vs post-update: 4.3 ± 3.6 vs 0.4 ± 0.8, 0.32 ± 0.70 vs 0.04 ± 0.20, and 8.6 ± 6.7 vs 1.6 ± 1.7, respectively). There was no significant decrease in the true positive rate (post-update vs pre-update: 95.0% vs 99.2%; P = 0.09) or the adenoma detection rate (post-update vs pre-update: 52.1% vs 49.3%; P = 0.87). CONCLUSIONS: The updated CADe can reduce FP without impairing polyp detection. A reduction in FP may help relieve the burden on endoscopists.

Deep learning system for true- and pseudo-invasion in colorectal polyps.

Over 15 million colonoscopies were performed yearly in North America, during which biopsies were taken for pathological examination to identify abnormalities. Distinguishing between true- and pseudo-invasion in colon polyps is critical in treatment planning. Surgical resection of the colon is often the treatment option for true invasion, whereas observation is recommended for pseudo-invasion. The task of identifying true- vs pseudo-invasion, however, could be highly challenging. There is no specialized software tool for this task, and no well-annotated dataset is available. In our work, we obtained (only) 150 whole-slide images (WSIs) from the London Health Science Centre. We built three deep neural networks representing different magnifications in WSIs, mimicking the workflow of pathologists. We also built an online tool for pathologists to annotate WSIs to train our deep neural networks. Results showed that our novel system classifies tissue types with 95.3% accuracy and differentiates true- and pseudo-invasions with 83.9% accuracy. The system's efficiency is comparable to an expert pathologist. Our system can also be easily adjusted to serve as a confirmatory or screening tool. Our system (available at http://ai4path.ca ) will lead to better, faster patient care and reduced healthcare costs.

Sporadic Polyps of the Colorectum.

Colorectal polyps are common, and their diagnosis and classification represent a major component of gastrointestinal pathology practice. The majority of colorectal polyps represent precursors of either the chromosomal instability or serrated neoplasia pathways to colorectal carcinoma. Accurate reporting of these polyps has major implications for surveillance and thus for cancer prevention. In this review, we discuss the key histologic features of the major colorectal polyps with a particular emphasis on diagnostic pitfalls and areas of contention.

Comparison of Right Colon Flat Polyp Detection Rate by Water Exchange Versus Water Immersion-pooled Results of Data File of 3 Published Reports.

GOALS: We tested the hypothesis that water exchange (WE) achieved a significantly higher right colon flat polyp detection rate (rFPDR) than water immersion (WI). BACKGROUND: Current endoscopy methods provide real-time morphology but not histopathology. Flat serrated polyps are difficult to find during colonoscopy. In 2022 2 studies reported that the serrated polyp detection rate (SPDR) significantly inversely predicted the development of interval cancers. In 2021 1 systemic review with meta-analysis showed that WE, but not WI increased SPDR. The relative contributions of WE and WI on rFPDR are unknown. STUDY: Individual patient data from 3 reports comparing air insufflation, WI, and WE were pooled. Multiple logistic regression analysis was used to assess the factors associated with a higher rFPDR. RESULTS: The pooled data showed that the rFPDR of air insufflation, WI, and WE were 15.4%, 14.1%, and 19.4% ( P =0.009), respectively. After adjusting for age and withdrawal time, multiple logistic regression analysis revealed that WE, when compared with WI, was significantly associated with a higher rFPDR (adjusted odds ratio[aOR]=1.53, P =0.002). Analysis of data on pathology and size were omitted to avoid duplicating our earlier publications. CONCLUSIONS: Significantly higher rFPDR was achieved by WE. Water exchange rather than WI merits consideration for use to maximize rFPDR. Removal of flat polyps, and by inference serrated polyps, ensures their optimal management to minimize the occurrence of interval cancers. The potential benefit of WE in maximizing SPDR and minimizing interval cancers deserves evaluation in long-term randomized controlled studies focused on flat polyps detection.

Impact of an artificial intelligence-aided endoscopic diagnosis system on improving endoscopy quality for trainees in colonoscopy: Prospective, randomized, multicenter study.

OBJECTIVE: This study was performed to evaluate whether the use of CAD EYE (Fujifilm, Tokyo, Japan) for colonoscopy improves colonoscopy quality in gastroenterology trainees. METHODS: The patients in this multicenter randomized controlled trial were divided into Group A (observation using CAD EYE) and Group B (standard observation). Six trainees performed colonoscopies using a back-to-back method in pairs with gastroenterology experts. The primary end-point was the trainees' adenoma detection rate (ADR), and the secondary end-points were the trainees' adenoma miss rate (AMR) and Assessment of Competency in Endoscopy (ACE) tool scores. Each trainee's learning curve was evaluated using a cumulative sum (CUSUM) control chart. RESULTS: We analyzed data for 231 patients (Group A, n = 113; Group B, n = 118). The ADR was not significantly different between the two groups. Group A had a significantly lower AMR (25.6% vs. 38.6%, P = 0.033) and number of missed adenomas per patient (0.5 vs. 0.9, P = 0.004) than Group B. Group A also had significantly higher ACE tool scores for pathology identification (2.26 vs. 2.07, P = 0.030) and interpretation and identification of pathology location (2.18 vs. 2.00, P = 0.038). For the CUSUM learning curve, Group A showed a trend toward a lower number of cases of missed multiple adenomas by the six trainees. CONCLUSION: CAD EYE did not improve ADR but decreased the AMR and improved the ability to accurately locate and identify colorectal adenomas. CAD EYE can be assumed to be beneficial for improving colonoscopy quality in gastroenterology trainees. TRIAL REGISTRATION: University Hospital Medical Information Network Clinical Trials Registry (UMIN000044031).

The Effect of Intervention for Improving Colonoscopy Quality Is Associated with the Personality Traits of Endoscopists.

Background/Aims: This study investigated whether the personality traits of endoscopists are associated with the effect of interventions for the improvement of colonoscopy quality. Methods: This prospective, multicenter, single-blind study was performed with 13 endoscopists in three health screening centers over a 12-month period. Quality indicators (QIs), including adenoma detection rate (ADR), polyp detection rate (PDR), and withdrawal time, were measured every 3 months. Consecutive interventions for the improvement of colonoscopy quality were conducted every 3 months, which included the personal notification of QIs, the in-group notification of QIs, and finally a targeted "quality education" session. The personality traits of each endoscopist were evaluated for perfectionism, fear of negative evaluation, and cognitive flexibility after the last QI assessment. Results: A total of 4,095 colonoscopies were evaluated to measure the QIs of the individual endoscopists for 12 months. The mean ADR, PDR, and withdrawal time of the 13 endoscopists were 32.3%, 47.7%, and 394 seconds at baseline and increased to 39.0%, 55.1%, and 430 seconds by the end of the study (p=0.003, p=0.006, and p=0.004, respectively). Among the three interventions, only quality education significantly improved QIs: ADR, 36.0% to 39.0% (odds ratio, 1.28; 95% confidence interval, 1.01 to 1.63). The improvement of ADR and PDR by education was significantly associated with perfectionism (r=0.617, p=0.033 and r=0.635, p=0.027, respectively) and fear of negative evaluation (r=0.704, p=0.011 and r=0.761, p=0.004, respectively). Conclusions: Education can improve colonoscopy quality, and its effect size is associated with an endoscopist's personal traits such as perfectionism and fear of negative evaluation (Clinical-Trials.gov Registry NCT03796169).