An Automatic Segmentation of Polyp in Colorectal Cancer Using U-Net

Jovita Relasha Lewis; Sameena Pathan; Preetham Kumar; Cifha Crecil Dias

doi:10.1007/978-3-032-14531-4_13

An Automatic Segmentation of Polyp in Colorectal Cancer Using U-Net

Jovita Relasha Lewis
, Sameena Pathan^*
, Preetham Kumar
, Cifha Crecil Dias

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Colorectal Cancer (CRC) is a main cause of cancer-associated deaths globally, with a significant number of victims due to late diagnosis. There is a correlation between colorectal polyps and the occurrence of CRC, highlighting the need for early intervention and diagnosis. The capability to segment the polyp accurately is critical as it would facilitate timely treatment and improve diagnosis accuracy. However, polyp segmentation presents challenges caused by different sizes, and shapes of the polyps, also in cases when they are concealed beneath the mucosal area. U-Net architecture is used to address these challenges to enhance the accuracy of the polyp identified. Our study utilized the Kvasir-Seg dataset. The U-Net model was trained on this data. The model demonstrated a loss of 0.225 and a validating accuracy of 0.9021. This advanced segmentation technique helps the gastroenterologist identify polyps with high accuracy and ultimately helps the patient with early detection and diagnosis.

Original language	English
Title of host publication	Machine Learning, Image Processing, Network Security and Data Sciences - 6th International Conference, MIND 2024, Revised Selected Papers
Editors	Chirag Modi, Veena Thenkanidiyoor, Gyanendra Kumar Verma, Ljiljana Brankovic
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	152-163
Number of pages	12
ISBN (Print)	9783032145307
DOIs	https://doi.org/10.1007/978-3-032-14531-4_13
Publication status	Published - 2026
Event	6th International Conference on Machine Learning, Image Processing, Network Security, and Data Sciences, MIND 2024 - Goa, India Duration: 20-12-2024 → 21-12-2024

Publication series

Name	Communications in Computer and Information Science
Volume	2736 CCIS
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	6th International Conference on Machine Learning, Image Processing, Network Security, and Data Sciences, MIND 2024
Country/Territory	India
City	Goa
Period	20-12-24 → 21-12-24

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

All Science Journal Classification (ASJC) codes

General Computer Science
General Mathematics

Access to Document

10.1007/978-3-032-14531-4_13

Cite this

Lewis, J. R., Pathan, S., Kumar, P., & Dias, C. C. (2026). An Automatic Segmentation of Polyp in Colorectal Cancer Using U-Net. In C. Modi, V. Thenkanidiyoor, G. K. Verma, & L. Brankovic (Eds.), Machine Learning, Image Processing, Network Security and Data Sciences - 6th International Conference, MIND 2024, Revised Selected Papers (pp. 152-163). (Communications in Computer and Information Science; Vol. 2736 CCIS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-032-14531-4_13

Lewis, Jovita Relasha ; Pathan, Sameena ; Kumar, Preetham et al. / An Automatic Segmentation of Polyp in Colorectal Cancer Using U-Net. Machine Learning, Image Processing, Network Security and Data Sciences - 6th International Conference, MIND 2024, Revised Selected Papers. editor / Chirag Modi ; Veena Thenkanidiyoor ; Gyanendra Kumar Verma ; Ljiljana Brankovic. Springer Science and Business Media Deutschland GmbH, 2026. pp. 152-163 (Communications in Computer and Information Science).

@inproceedings{3286f2525bbe495aa4759f62ea016101,

title = "An Automatic Segmentation of Polyp in Colorectal Cancer Using U-Net",

abstract = "Colorectal Cancer (CRC) is a main cause of cancer-associated deaths globally, with a significant number of victims due to late diagnosis. There is a correlation between colorectal polyps and the occurrence of CRC, highlighting the need for early intervention and diagnosis. The capability to segment the polyp accurately is critical as it would facilitate timely treatment and improve diagnosis accuracy. However, polyp segmentation presents challenges caused by different sizes, and shapes of the polyps, also in cases when they are concealed beneath the mucosal area. U-Net architecture is used to address these challenges to enhance the accuracy of the polyp identified. Our study utilized the Kvasir-Seg dataset. The U-Net model was trained on this data. The model demonstrated a loss of 0.225 and a validating accuracy of 0.9021. This advanced segmentation technique helps the gastroenterologist identify polyps with high accuracy and ultimately helps the patient with early detection and diagnosis.",

author = "Lewis, \{Jovita Relasha\} and Sameena Pathan and Preetham Kumar and Dias, \{Cifha Crecil\}",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.; 6th International Conference on Machine Learning, Image Processing, Network Security, and Data Sciences, MIND 2024 ; Conference date: 20-12-2024 Through 21-12-2024",

year = "2026",

doi = "10.1007/978-3-032-14531-4\_13",

language = "English",

isbn = "9783032145307",

series = "Communications in Computer and Information Science",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "152--163",

editor = "Chirag Modi and Veena Thenkanidiyoor and Verma, \{Gyanendra Kumar\} and Ljiljana Brankovic",

booktitle = "Machine Learning, Image Processing, Network Security and Data Sciences - 6th International Conference, MIND 2024, Revised Selected Papers",

address = "Germany",

}

Lewis, JR, Pathan, S , Kumar, P & Dias, CC 2026, An Automatic Segmentation of Polyp in Colorectal Cancer Using U-Net. in C Modi, V Thenkanidiyoor, GK Verma & L Brankovic (eds), Machine Learning, Image Processing, Network Security and Data Sciences - 6th International Conference, MIND 2024, Revised Selected Papers. Communications in Computer and Information Science, vol. 2736 CCIS, Springer Science and Business Media Deutschland GmbH, pp. 152-163, 6th International Conference on Machine Learning, Image Processing, Network Security, and Data Sciences, MIND 2024, Goa, India, 20-12-24. https://doi.org/10.1007/978-3-032-14531-4_13

An Automatic Segmentation of Polyp in Colorectal Cancer Using U-Net. / Lewis, Jovita Relasha; Pathan, Sameena ; Kumar, Preetham et al.
Machine Learning, Image Processing, Network Security and Data Sciences - 6th International Conference, MIND 2024, Revised Selected Papers. ed. / Chirag Modi; Veena Thenkanidiyoor; Gyanendra Kumar Verma; Ljiljana Brankovic. Springer Science and Business Media Deutschland GmbH, 2026. p. 152-163 (Communications in Computer and Information Science; Vol. 2736 CCIS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - An Automatic Segmentation of Polyp in Colorectal Cancer Using U-Net

AU - Lewis, Jovita Relasha

AU - Pathan, Sameena

AU - Kumar, Preetham

AU - Dias, Cifha Crecil

N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.

PY - 2026

Y1 - 2026

N2 - Colorectal Cancer (CRC) is a main cause of cancer-associated deaths globally, with a significant number of victims due to late diagnosis. There is a correlation between colorectal polyps and the occurrence of CRC, highlighting the need for early intervention and diagnosis. The capability to segment the polyp accurately is critical as it would facilitate timely treatment and improve diagnosis accuracy. However, polyp segmentation presents challenges caused by different sizes, and shapes of the polyps, also in cases when they are concealed beneath the mucosal area. U-Net architecture is used to address these challenges to enhance the accuracy of the polyp identified. Our study utilized the Kvasir-Seg dataset. The U-Net model was trained on this data. The model demonstrated a loss of 0.225 and a validating accuracy of 0.9021. This advanced segmentation technique helps the gastroenterologist identify polyps with high accuracy and ultimately helps the patient with early detection and diagnosis.

AB - Colorectal Cancer (CRC) is a main cause of cancer-associated deaths globally, with a significant number of victims due to late diagnosis. There is a correlation between colorectal polyps and the occurrence of CRC, highlighting the need for early intervention and diagnosis. The capability to segment the polyp accurately is critical as it would facilitate timely treatment and improve diagnosis accuracy. However, polyp segmentation presents challenges caused by different sizes, and shapes of the polyps, also in cases when they are concealed beneath the mucosal area. U-Net architecture is used to address these challenges to enhance the accuracy of the polyp identified. Our study utilized the Kvasir-Seg dataset. The U-Net model was trained on this data. The model demonstrated a loss of 0.225 and a validating accuracy of 0.9021. This advanced segmentation technique helps the gastroenterologist identify polyps with high accuracy and ultimately helps the patient with early detection and diagnosis.

UR - https://www.scopus.com/pages/publications/105029421641

UR - https://www.scopus.com/pages/publications/105029421641#tab=citedBy

U2 - 10.1007/978-3-032-14531-4_13

DO - 10.1007/978-3-032-14531-4_13

M3 - Conference contribution

AN - SCOPUS:105029421641

SN - 9783032145307

T3 - Communications in Computer and Information Science

SP - 152

EP - 163

BT - Machine Learning, Image Processing, Network Security and Data Sciences - 6th International Conference, MIND 2024, Revised Selected Papers

A2 - Modi, Chirag

A2 - Thenkanidiyoor, Veena

A2 - Verma, Gyanendra Kumar

A2 - Brankovic, Ljiljana

PB - Springer Science and Business Media Deutschland GmbH

T2 - 6th International Conference on Machine Learning, Image Processing, Network Security, and Data Sciences, MIND 2024

Y2 - 20 December 2024 through 21 December 2024

ER -

Lewis JR, Pathan S , Kumar P , Dias CC. An Automatic Segmentation of Polyp in Colorectal Cancer Using U-Net. In Modi C, Thenkanidiyoor V, Verma GK, Brankovic L, editors, Machine Learning, Image Processing, Network Security and Data Sciences - 6th International Conference, MIND 2024, Revised Selected Papers. Springer Science and Business Media Deutschland GmbH. 2026. p. 152-163. (Communications in Computer and Information Science). doi: 10.1007/978-3-032-14531-4_13