CT colonography in cancer detection: methods and results

An outlier removal method based on PCA-DBSCAN for blood-SERS data analysis

Surface-enhanced Raman spectroscopy (SERS) has shown promising potential in cancer screening. In practical applications, Raman spectra are often affected by deviations from the spectrometer, changes in measurement environments, and anomalies in spectrum characteristic peak intensities due to improper sample storage. Previous research has overlooked the presence of outliers in categorical data, leading to significant impacts on model learning outcomes. In this study, we propose a novel method, called Principal Component Analysis and Density Based Spatial Clustering of Applications with Noise (PCA-DBSCAN) to effectively remove outliers. This method employs dimensionality reduction and spectral data clustering to identify and remove outliers. The PCA-DBSCAN method introduces adjustable parameters (Eps and MinPts) to control the clustering effect. The effectiveness of the proposed PCA-DBSCAN method is verified through modeling on outlier-removed datasets. Further refinement of the machine learning model and PCA-DBSCAN parameters resulted in the best cancer screening model, achieving 97.41% macro-average recall and 97.74% macro-average F1-score. This paper introduces a new outlier removal method that significantly improves the performance of the SERS cancer screening model. Moreover, the proposed method serves as inspiration for outlier detection in other fields, such as biomedical research, environmental monitoring, manufacturing, quality control, and hazard prediction.

Laparoscopic compatible device incorporating inductive proximity sensors for precise detection of gastric and colorectal small tumors

The accurate localization of small tumors of the digestive tract is of paramount importance in surgical oncology because it dictates the limits of resection and the extent of lymph node dissection. In this view, we have designed and fabricated a highly efficient sensing laparoscopic instrument focused on precise non-invasive extralumenal intraoperative detection of small colorectal or gastric tumors. The equipment is fully adapted for laparoscopic surgery and consists of an inductive proximity sensor encapsulated into a watertight stainless-steel case that is connected to an electronic functional block dimensionally scaled-down by the desired form and size for optimal surgical manipulation. The sensor-case unit and the electronic block are coupled together using a modular system which allows disconnection of the latter and sterilization by autoclavation of the former, followed by swift plugging of the electronic block just before surgery in a sterile-controlled environment. The instrument works in tandem with a modified endoscopic hemostatic clip which is attached endoscopically, before surgery, in the mucosa proximal and distal to the tumor. By scanning the serosal side of the digestive organ during the laparoscopic surgical procedure, the detector senses the modified clip and thus pinpoints to the location of the tumor. Additional engineering of the standard endoscopic hemostatic clips by coating them with various combinations of metallic alloys of Cu and Zn was necessary to improve the detection range and sensitivity without compromising on their functionality. The clips were also covered with nanometric layers of Au to ensure their biocompatibility. The ex-vivo dry-lab experiments showed a satisfactory detection distance which was later confirmed in ex-vivo wet-lab experiments on animal organs and human surgical specimens.

CT colonography for synchronous colorectal lesions in patients with colorectal cancer: initial experience

AIM

To assess accuracy of CT colonography (CTC) in identifying synchronous lesions in patients with colorectal carcinoma.

METHODS

This study included 174 consecutive patients undergoing CTC as part of staging or primary investigation where a colorectal cancer was diagnosed between 2004 and 2007. Prone unenhanced and portal phase enhanced supine series with air or CO(2) distension were acquired using 4- or 16-slice CT (Toshiba) and read by 2D +/- 3D formats. Synchronous lesions were classified according to American College of Radiology's (ACR) polyp classification. Segmental gold standard was flexible sigmoidoscopy/colonoscopy within 1 year and/or histology of colonic resection supplemented by follow-up. Nine patients without gold standard were excluded. Sensitivity, specificity and accuracy were calculated on a per polyp, per patient and per segment basis and discrepancies analysed.

RESULTS

Direct comparable data were available for 764/990 colonic segments from 165 patients. Of 41 (C2-C4) synchronous lesions on "gold standard", 33 were correctly identified on virtual colonoscopy (VC), overall per polyp sensitivity was 80.5%, with detection rates of 20/24 C3 (83.3%) and 3/3 C4 (100%) with per patient and per segment specificity of 95.4% and 99.2%, respectively.

CONCLUSION

CTC is an accurate technique to assess for significant synchronous lesions in patients with colorectal cancer and is applicable for total pre-operative colonic visualisation.

Patient satisfaction with colonoscopy: a literature review and pilot study

BACKGROUND

Current guidelines recommend that colonoscopic colorectal cancer screening be undertaken every 10 years after the age of 50 years. However, because the procedure does not meet criteria that promote screening uptake, patient satisfaction with colonoscopy may encourage repeat screening.

OBJECTIVE

To systematically review the literature and conduct a pilot study of patient satisfaction with the colonoscopy experience.

METHODS

All cohort studies from January 1997 to August 2008 in the MEDLINE database that measured either patient satisfaction with colonoscopy, patient willingness to return for colonoscopy under the same conditions or patient preference for colonoscopy compared with other large bowel procedures were identified. The search was supplemented by journal citation lists in the retrieved articles.

RESULTS

Of the 29 studies identified, 15 met the inclusion criteria. Consistently, the vast majority of patients (approximately 95%) were very satisfied with their colonoscopy experience. Patient satisfaction was similar for screening and nonscreening colonoscopy. Patient willingness to return for the procedure ranged from 73% to 100%. Of the five studies that examined modality preference, three studies reported the majority of patients preferred colonography to colonoscopy and two studies reported the reverse. Our pilot study findings mirrored those of other studies that were conducted in the United States. The major limitation of the included studies was that patients who were most dissatisfied may have gone elsewhere to have their colonoscopy.

CONCLUSIONS

Patients were very satisfied with colonoscopy. The majority were willing to return for repeat testing under the same conditions, and colonoscopy was not preferred over other modalities. However, studies were limited by methodological shortcomings.

Computed tomography colonography in acromegaly

INTRODUCTION AND AIM

Acromegalic patients have an increased risk for the development of colorectal cancer. For this reason, since 1996, screening colonoscopy has been recommended in all patients with acromegaly. The aim of our study was to assess the feasibility and to evaluate the results of computed tomography (CT)-colonography in acromegaly.

PATIENTS AND METHODS

We examined 23 acromegalic patients with no history of colorectal cancer (11 females and 12 males; age range 18-79 yr; disease duration range 1-15 yr) with CT-colonography. Twenty of them underwent traditional colonoscopy after the CT-colonography.

RESULTS

CT-colonography examination results were adequate in 17 of 23 cases (73%). CT-colonography found 12 polyps in eight patients, 95% confirmed by traditional colonoscopy. One polyp was a sigmoid cancer, and the diagnosis was confirmed at surgery. There were no polyps found by traditional colonoscopy that CT-colonography was not able to identify. The lesions were located in right colon (two), transversum (three), left colon (five), and sigmoid colon (two). Patient acceptance of the technique was good in 65%, medium in 20%, and poor in 15%.

CONCLUSION

For the first time we have demonstrated that CT-colonography has the potential ability to replace traditional colonoscopy in acromegalic patients. CT-colonography could be used as a screening modality for colon cancer in acromegaly.