scholarly journals CT Radiomics in Colorectal Cancer: Detection of KRAS Mutation Using Texture Analysis and Machine Learning

2020 ◽  
Vol 10 (18) ◽  
pp. 6214
Author(s):  
Víctor González-Castro ◽  
Eva Cernadas ◽  
Emilio Huelga ◽  
Manuel Fernández-Delgado ◽  
Jacobo Porto ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Texture Analysis ◽  
Cancer Detection ◽  
Kras Mutation ◽  
Texture Feature ◽  
University Hospital ◽  
Support Vector ◽  
Tomography Image ◽  
Computed Tomography Image ◽  
The University

In this work, by using descriptive techniques, the characteristics of the texture of the CT (computed tomography) image of patients with colorectal cancer were extracted and, subsequently, classified in KRAS+ or KRAS-. This was accomplished by using different classifiers, such as Support Vector Machine (SVM), Grading Boosting Machine (GBM), Neural Networks (NNET), and Random Forest (RF). Texture analysis can provide a quantitative assessment of tumour heterogeneity by analysing both the distribution and relationship between the pixels in the image. The objective of this research is to demonstrate that CT-based Radiomics can predict the presence of mutation in the KRAS gene in colorectal cancer. This is a retrospective study, with 47 patients from the University Hospital, with a confirmatory pathological analysis of KRAS mutation. The highest accuracy and kappa achieved were 83% and 64.7%, respectively, with a sensitivity of 88.9% and a specificity of 75.0%, achieved by the NNET classifier using the texture feature vectors combining wavelet transform and Haralick coefficients. The fact of being able to identify the genetic expression of a tumour without having to perform either a biopsy or a genetic test is a great advantage, because it prevents invasive procedures that involve complications and may present biases in the sample. As well, it leads towards a more personalized and effective treatment.

Robotic Colonoscopy: Comparative Analysis of Costs Compared to Painless Conventional Colonoscopy

2020 ◽  
Author(s):  
Rosi Sicuro ◽  
Emanuele Tumino ◽  
Christian Lambiase ◽  
Domenica Mamone
Keyword(s):  
Colorectal Cancer ◽  
Diagnostic Accuracy ◽  
Impact Analysis ◽  
Early Stage ◽  
Health Technology ◽  
University Hospital ◽  
Time Interval ◽  
Diagnostic Colonoscopy ◽  
High Diagnostic Accuracy ◽  
The University

Abstract BackgroundAmong the various causes of death, colorectal carcinoma represents the second highest cause in frequency both in men and in women. A colorectal cancer is diagnosed every 3.5 minutes and a person dies of colorectal cancer every 9 minutes. In 2018, in Italy were recorded around 51.000 new cases, with a mortality rate of over 18.000 deaths. There is clear evidence demonstrating that the identification and treatment of cancer at an early stage positively influence the reduction in mortality. Colonoscopy is the most effective technique used to identify and remove polyps, thus avoiding the costs related to surgical treatment and hospitalization.The purpose of this study is to evaluate the costs of the system Endotics for robotic colonoscopy as an alternative to conventional diagnostic colonoscopy performed under anaesthesia at the University Hospital of Pisa.MethodsThe cost analysis was developed according to the Budget Impact Analysis method application, an essential and complementary part of the Health Technology Assessment evaluation, which has the main purpose of assessing the financial sustainability of new health technology, estimating the consequences of its use and diffusion in a specific context characterized by the limited availability of resources. The observation period covered a time interval of 3 months, during which an average of 43 colonoscopic procedures per day was performed and mapped and a total of 23 colonoscopes were used. ResultsOverall, the work done has allowed identifying the cost of a conventional painless diagnostic colonoscopy performed in our institution which amounts to € 426.25. The valuation of the costs of the robotic colonoscopy amounted overall to € 441.25.ConclusionsThe ideal procedure to diagnose a colon disease should be safe, well-tolerated, possibly non-invasive, with high diagnostic accuracy and, not least, cost-effectiveness. The results of this study suggest that in the University Hospital of Pisa the costs related to robotic colonoscopy performed with the Endotics system are superimposable to those of conventional painless colonoscopy, reducing the overall risk associated with the colonoscopic procedure maintaining a high diagnostic accuracy with a greater tolerability by the patient, thus pushing the colonoscopy towards “the ideal procedure”.

scholarly journals Colorectal Cancer Screening Programme: Algarve Preliminary data

2019 ◽  
Vol 29 (Supplement_4) ◽  
Author(s):  
M Correia ◽  
M Cancela ◽  
P Caldeira ◽  
J Morera ◽  
C Cardoso ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Cancer Screening ◽  
Early Diagnosis ◽  
Colorectal Cancer Screening ◽  
Cancer Detection ◽  
Burden Of Disease ◽  
Public Awareness ◽  
Target Population ◽  
University Hospital ◽  
Invasive Treatment

Abstract Problem The burden of disease for oncological problems is significant in Portugal. Population-based screenings, such as colorectal cancer screening (CCS) with faecal occult blood tests (FIT) is a priority program to reduce the burden of disease and a cost-effectiveness measure. Description problem Algarve CCS started in 2017,to early detect colorectal cancer reducing morbidity and mortality. Screening strategy to detect and remove cancer precursors (adenomas) and early cancer detection with less invasive treatment using a non-invasive stool test was selected. A quantitative faecal immunochemical test for haemoglobin (FIT) with higher sensitivity for adenoma and cancer, specific for human globin with no dietary restriction, with a single sample screening is considered a test with higher participation. Target population all gender from 50 to 75 are invited to participate in the CCS by letter. Screening interval is 2 years and FIT cut-off- 117 (ng Hb/mL). Software was used to manage all CCS. Health units distribute the test and a flyer. All FIT-positive were invited to do a full colonoscopy at University Hospital. Results A preliminary descriptive analysis (2017/2019) was made. After exclusions 25017 of target population participate 34,4%(n = 8594),27.5%(n = 6883) submitted it for analysis, 3.9% (n = 266) had positive FIT and were invited for a full colonoscopy with 57.5% (n = 153)participation rate. Cancer detection was 0.9‰ and adenomas detection rate was 9.9%. FIT positive with colonoscopy 3.9% (n = 6) had carcinoma and 44.4% (n = 68) had adenoma. Lessons Some operational changes were made to improve participants and professionals adherence to CCS. Pathology observed in the positive cases, reinforce the need to implement strategies to increase literacy, raise public awareness, keep the screening test and improve accessibility of health facilities. Key messages The importance of a colorectal cancer early diagnosis prevent aggressive interventions in a precancerous phase. The importance of a colorectal cancer early diagnosis improves survival and population quality of life.

MULTI-RESOLUTION CROWD DENSITY ESTIMATION BASED ON TEXTURE ANALYSIS AND LEARNING FROM DEMONSTRATION

2007 ◽  
Vol 04 (01) ◽  
pp. 1-14 ◽  
Author(s):  
GUOYUAN LIANG ◽  
KA KEUNG LEE ◽  
YANGSHENG XU
Keyword(s):  
Texture Analysis ◽  
Density Estimation ◽  
Texture Feature ◽  
Texture Features ◽  
Input Image ◽  
Support Vector ◽  
Surveillance Systems ◽  
Regression Problem ◽  
Crowd Density Estimation ◽  
Crowd Density

Crowd density estimation is very important for intelligent surveillance systems in public places. This paper presents an automatic method of estimating crowd density using texture analysis and machine learning. First the crowd scene is modeled as a series of multi-resolution image cells based on perspective projection. The cell size is normalized to obtain a uniform representation of texture features. Then the feature vectors of textures are extracted from each input image cell and the support vector machine (SVM) method is utilized to solve the regression problem for calculating the crowd density. In order to diminish the instability of texture feature measurements, a technique of searching the extrema in the Harris–Laplacian space is applied. Finally, the SVM method is used again to detect some abnormal situations caused by the changes in density distribution. Experiments on real crowd videos show the effectiveness of the proposed system.

scholarly journals The impact of COVID-19 epidemiological restriction guidelines measures in a Croatian tertiary colorectal cancer center

2020 ◽  
Vol 48 (2-3) ◽  
pp. 43-46
Author(s):  
Iva Kirac ◽  
◽  
Zvonimir Misir ◽  
Vesna Vorih ◽  
Loris Ćurt ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Colorectal Surgery ◽  
University Hospital ◽  
Cancer Center ◽  
Solid Cancer ◽  
Everyday Activities ◽  
The One ◽  
The University ◽  
The Impact ◽  
The Given

Background: In the past six months, Croatia faced a short lockdown and a slow return to most hospitals’ everyday activities. During the lockdown, our center, as a part of the University Hospital Centre specialized for solid cancer, was enabled to maintain most of the routine practices with the one-month colonoscopy exception. Aim: To determine the oscillation in the number of endoscopies and colorectal surgery for 13 months (six months pre and post COVID-19 lockdown). Materials and methods: From August 1st, 2019, until August 31st 2020, the hospital analytics determine the number of colonoscopies, screening colonoscopies, and surgeries. Results: During the given period number of detected and operated colorectal cancers was stable, except for April, when we mostly did not perform colonoscopies. Conclusion: We maintained a pre-COVID-19 pace in colorectal cancer treatment, colonoscopies, and colorectal surgery after epidemiological guidelines for colonoscopies and colorectal surgery were applied, owing to the relatively stable overall epidemiological situation.

Sign in / Sign up

Export Citation Format

Share Document