Artificial intelligence and machine learning for early detection and diagnosis of colorectal cancer in sub-Saharan Africa

Akbar K Waljee^{1

2

3

4}, Eileen M Weinheimer-Haus^{2

3

4}, Amina Abubakar⁵, Anthony K Ngugi⁶, Geoffrey H Siwo^{2

3

7

8}, Gifty Kwakye⁹, Amit G Singal^{10

11}, Arvind Rao^{4

12

13

14}, Sameer D Saini^{15

2}, Andrew J Read^{2

4}, Jessica A Baker^{15

3

4}, Ulysses Balis¹⁶, Christopher K Opio¹⁷, Ji Zhu^{3

4

18}, Mansoor N Saleh^{19

20}

Affiliations

¹ Veterans Affairs Center for Clinical Management Research, Ann Arbor, Michigan, USA awaljee@med.umich.edu.
² Department of Internal Medicine, Division of Gastroenterology, University of Michigan, Ann Arbor, Michigan, USA.
³ Center for Global Health Equity, University of Michigan, Ann Arbor, Michigan, USA.
⁴ Michigan Integrated Center for Health Analytics and Medical Prediction (MiCHAMP), University of Michigan, Ann Arbor, Michigan, USA.
⁵ Institute for Human Development, The Aga Khan University, Nairobi, Kenya.
⁶ Department of Population Health, The Aga Khan University, Nairobi, Kenya.
⁷ Eck Institute for Global Health, University of Notre Dame, South Bend, Indiana, USA.
⁸ Center for Research Computing, University of Notre Dame, South Bend, Indiana, USA.
⁹ Department of Surgery, Division of Colorectal Surgery, University of Michigan, Ann Arbor, Michigan, USA.
¹⁰ Harold C. Simmons Comprehensive Cancer Center, The University of Texas Southwestern Medical Center, Dallas, Texas, USA.
¹¹ Department of Internal Medicine, Division of Digestive and Liver Diseases, The University of Texas Southwestern Medical Center, Dallas, Texas, USA.
¹² Department of Computational Medicine and Bioinformatics, University of Michigan, Ann Arbor, Michigan, USA.
¹³ Department of Biomedical Engineering, University of Michigan, Ann Arbor, Michigan, USA.
¹⁴ Department of Radiation Oncology, University of Michigan, Ann Arbor, Michigan, USA.
¹⁵ Veterans Affairs Center for Clinical Management Research, Ann Arbor, Michigan, USA.
¹⁶ Department of Pathology, University of Michigan Health System, Ann Arbor, Michigan, USA.
¹⁷ Department of Medicine, Aga Khan University Hospital Nairobi, Nairobi, Kenya.
¹⁸ Department of Statistics, University of Michigan, Ann Arbor, Michigan, USA.
¹⁹ O'Neal Comprehensive Cancer Center, The University of Alabama at Birmingham, Birmingham, Alabama, USA.
²⁰ Department of Hematology-Oncology, Aga Khan University Hospital Nairobi, Nairobi, Kenya.

PMID: 35418482
PMCID: PMC9177787
DOI: 10.1136/gutjnl-2022-327211

Review

Artificial intelligence and machine learning for early detection and diagnosis of colorectal cancer in sub-Saharan Africa

Akbar K Waljee et al. Gut. 2022 Jul.

. 2022 Jul;71(7):1259-1265.

doi: 10.1136/gutjnl-2022-327211. Epub 2022 Apr 13.

Authors

Affiliations

¹ Veterans Affairs Center for Clinical Management Research, Ann Arbor, Michigan, USA awaljee@med.umich.edu.
² Department of Internal Medicine, Division of Gastroenterology, University of Michigan, Ann Arbor, Michigan, USA.
³ Center for Global Health Equity, University of Michigan, Ann Arbor, Michigan, USA.
⁴ Michigan Integrated Center for Health Analytics and Medical Prediction (MiCHAMP), University of Michigan, Ann Arbor, Michigan, USA.
⁵ Institute for Human Development, The Aga Khan University, Nairobi, Kenya.
⁶ Department of Population Health, The Aga Khan University, Nairobi, Kenya.
⁷ Eck Institute for Global Health, University of Notre Dame, South Bend, Indiana, USA.
⁸ Center for Research Computing, University of Notre Dame, South Bend, Indiana, USA.
⁹ Department of Surgery, Division of Colorectal Surgery, University of Michigan, Ann Arbor, Michigan, USA.
¹⁰ Harold C. Simmons Comprehensive Cancer Center, The University of Texas Southwestern Medical Center, Dallas, Texas, USA.
¹¹ Department of Internal Medicine, Division of Digestive and Liver Diseases, The University of Texas Southwestern Medical Center, Dallas, Texas, USA.
¹² Department of Computational Medicine and Bioinformatics, University of Michigan, Ann Arbor, Michigan, USA.
¹³ Department of Biomedical Engineering, University of Michigan, Ann Arbor, Michigan, USA.
¹⁴ Department of Radiation Oncology, University of Michigan, Ann Arbor, Michigan, USA.
¹⁵ Veterans Affairs Center for Clinical Management Research, Ann Arbor, Michigan, USA.
¹⁶ Department of Pathology, University of Michigan Health System, Ann Arbor, Michigan, USA.
¹⁷ Department of Medicine, Aga Khan University Hospital Nairobi, Nairobi, Kenya.
¹⁸ Department of Statistics, University of Michigan, Ann Arbor, Michigan, USA.
¹⁹ O'Neal Comprehensive Cancer Center, The University of Alabama at Birmingham, Birmingham, Alabama, USA.
²⁰ Department of Hematology-Oncology, Aga Khan University Hospital Nairobi, Nairobi, Kenya.

PMID: 35418482
PMCID: PMC9177787
DOI: 10.1136/gutjnl-2022-327211

No abstract available

Keywords: Artifiical Intelligence; COLORECTAL CANCER; SURVEILLANCE.

PubMed Disclaimer

Conflict of interest statement

Competing interests: AGS has consulted for and received research funding from Exact Sciences. AR serves as member for Voxel Analytics and consults for Genophyll and Pact&Health. GHS is a founder of Anza Biotechnologies.

Figures

**Figure 1**
Identification of colon cancer in a digital H&E-stained tissue section of colonic adenocarcinoma. (A) image of colon cancer from a digital slide. (B) a vector was created to identify only the malignant glands, and (C) an additional vector was created to recognise only the stroma. (D) Boolean logic was used to determine the malignant glands, and the stroma was subtracted out. This approach could assist pathologists in identifying small foci of invasive glands or small foci of tumour present in blood and lymphatic vessels, which might be otherwise overlooked. Figure copyright Hipp *et al*, licensed under CC-BY 2.0 (https://creativecommons.org/licenses/by/2.0/).

**Figure 2**
Summary of analysis workflow for identifying histological determinants of malignant transformation and disease grade. Step 3 uses a support vector machine classifier, but any classifier can be used (eg, random forest). Figure copyright Powell *et al*, licensed under CC-BY 2.0 (https://creativecommons.org/licenses/by/2.0/). BIC, Bayesian information criterion; OS, overall survival; SVM, support vector machine.

**Figure 3**
Examples of challenges and opportunities for leveraging AI-based approaches in sub-Saharan Africa. AI, artificial intelligence; CAB, community advisory board; CBO, community-based organisation; NGO, non-governmental organisations.

**Figure 4**
Depiction of the Harnessing Data Science for Health Discovery and Innovation in Africa (DS-I Africa) programme and exemplar research hub. (A) The four main initiatives are: (1) Research hubs will apply novel approaches to data analysis and artificial intelligence to address critical health issues in Africa. (2) Open data science platform and coordinating centre will provide a flexible, scalable platform for the DS-I Africa researchers to find and access data, select tools and workflows, and run analyses through collaborative workspaces. It will also deliver the organisational framework for the direction and management of the initiative’s common activities; (3) Research training programmes will create multi-tiered curricula to build skills in foundational health data science, with options ranging from master’s and doctoral degree tracks to postdoctoral training and faculty development; and (4) The ethical, legal and social implications (ELSI) projects will address data science issues that present challenges in Africa such as data privacy and ownership, cybersecurity and sensitivities concerning the use of geospatial information for research or public health surveillance. (B) Led by the Aga Khan University—East Africa, Kenya Medical Research Institute-Wellcome Trust Research Programme, and the University of Michigan, the research hub will implement two research projects around maternal, newborn and child health as well as mental health, which will be supported by three cores: Admin core, Data Management and Analysis Core (DMAC) and Dissemination and Sustainability Core (DSC). The Admin Core will lead the UZIMA-DS researchhub, fostering synergy and integration of all hub components and partnerships and facilitating participation in DS-I cross-consortium activities. The DMAC will employ FAIR (Findable, Accessible, Interoperable, Reusable) principles to support the hub’s data ecosystem through data governance, facilitating data analytics within the projects, and fostering data sharing and interoperability throughout the greater DS-I Africa consortium. The DSC will promote engagement with stakeholders to identify sustainable model dissemination pathways into target communities. Through multisectoral partnerships with government, healthcare and non-profit sectors, the core will: facilitate the development of best practices and policies with stakeholders using data-driven approaches to inform guidelines; and promote engagement with private sectors to explore sustainable commercialisation opportunities and pathways. UZIMA-DS, UtiliZing Health Information for Meaningful Impact in East Africa Through Data Science.

See this image and copyright information in PMC

References

1. Sung H, Ferlay J, Siegel RL, et al. . Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2021;71:209–49. 10.3322/caac.21660 - DOI - PubMed
1. Parker RK, Mwachiro MM, Ranketi SS, et al. . Curative surgery improves survival for colorectal cancer in rural Kenya. World J Surg 2020;44:30–6. 10.1007/s00268-019-05234-1 - DOI - PubMed
1. Saluja S, Alatise OI, Adewale A, et al. . A comparison of colorectal cancer in Nigerian and North American patients: is the cancer biology different? Surgery 2014;156:305–10. 10.1016/j.surg.2014.03.036 - DOI - PMC - PubMed
1. Asombang AW, Madsen R, Simuyandi M, et al. . Descriptive analysis of colorectal cancer in Zambia, southern Africa using the National cancer disease Hospital database. Pan Afr Med J 2018;30:248. 10.11604/pamj.2018.30.248.12464 - DOI - PMC - PubMed
1. Gullickson C, Goodman M, Joko-Fru YW, et al. . Colorectal cancer survival in sub-Saharan Africa by age, stage at diagnosis and human development index: a population-based registry study. Int J Cancer 2021;149:1553–63. 10.1002/ijc.33715 - DOI - PubMed

Publication types

Actions
Actions

MeSH terms

Actions
Actions
Actions
Actions
Actions
Actions
Actions

Grants and funding

U54 TW012089/TW/FIC NIH HHS/United States

LinkOut - more resources

Full Text Sources
Medical
- MedlinePlus Health Information

Save citation to file

Email citation

Add to Collections

Add to My Bibliography

Your saved search

Create a file for external citation management software

Your RSS Feed

Artificial intelligence and machine learning for early detection and diagnosis of colorectal cancer in sub-Saharan Africa

Affiliations

Artificial intelligence and machine learning for early detection and diagnosis of colorectal cancer in sub-Saharan Africa

Authors

Affiliations

Conflict of interest statement

Figures

References

Publication types

MeSH terms

Grants and funding

LinkOut - more resources

Full Text Sources

Medical