This document provides an overview of research ethics and ethical principles. It discusses the different types of ethics including meta-ethics, prescriptive ethics, descriptive ethics, and applied ethics. The key principles of research ethics are described, including minimizing harm, obtaining informed consent, protecting anonymity and confidentiality, avoiding misleading practices, and providing the right to withdraw. Ethical issues in medical research and approaches to resolving ethical dilemmas are also summarized. Research misconduct involving fabrication, falsification, or plagiarism is defined.

1. i
UNIVERSITY OF GONDAR
College of Medicine and Health Science
IPH PhD School
Department of Public Health
Course Title: Ethics in Health Research
Course code: PuHe803
Seminar Topic: Ethics of research

2. ii
Table of Contents
Acknowledgment ...........................................................................Error! Bookmark not defined.
List of tables...................................................................................................................................iii
Acronym ........................................................................................................................................ iv
Introduction to Ethics...................................................................................................................... 1
Types of Ethics ............................................................................................................................... 1
Ethics in Research........................................................................................................................... 2
Objectives of research ethics....................................................................................................... 1
Principles of Research Ethics...................................................................................................... 1
Principles of ethics in medical research.......................................................................................... 5
Ethical Dilemma ............................................................................................................................. 6
Ethical Dilemmas and Morals..................................................................................................... 7
Approach to Solve an Ethical Dilemma.......................................................................................... 7
Research Misconduct...................................................................................................................... 7
Ethical issues in my dissertation research....................................................................................... 8
Ethical principles in the use of deep learning in healthcare............................................................ 8
Ethical dilemmas in deep learning .............................................................................................. 1
Ethical dilemma in Breast cancer................................................................................................ 1
Research Misconducts in Deep learning:.................................................................................... 1
Research Misconducts in Breast cancer:..................................................................................... 1
Mitigation strategies for research dilemma and misconduct....................................................... 1
Ethical Challenges and Mitigation strategies in my dissertation ................................................ 2
Reference ........................................................................................................................................ 3

3. iii
List of tables
Table-1: Principles of Research Ethics---------------------------------------------------------------------3
Table-2: Ethical Challenges and Mitigation strategies -------------------------------------------------11

4. iv
Acronym
 AI --------------------------------- Artificial Intelligence
 ML -------------------------------- Machine Learning

5. 1
Introduction to Ethics
Ethics is a branch of philosophy that is concerned with human conduct, more specifically the
behavior of individuals in society. Ethics derived from the Greek word "ethos," which means
"custom," "habit," or “way of living”. Ethics examines the rational justification for our moral
judgments, it studies what is morally right or wrong, good or bad, just or unjust. Ethics help to
make moral decisions and understand, obligations and responsibilities to others [1].
Types of Ethics
Ethics is mainly divided into four main branches. They are as follows [2]:
1. Meta-Ethics: (Ethics about Ethics)
2. Prescriptive Ethics: (Normative Ethics)
3. Descriptive Ethics: (Comparative Ethics)
4. Applied Ethics
Meta-Ethics (Types of Ethics)
Meta-ethics can be defined as a branch of ethics that is concerned with the study of the nature of
ethics. It analyzes the meaning when we use words like good, bad, right, and wrong. Meta-Ethics
is more about philosophy in nature as it deals with the nature of ethics and morality. It
investigates where our moral and ethical principles come from and what is the meaning behind
using them.
 Deals with questions like what is meant by being right? OR what is meant by being
wrong?
 Deals with the definition of right and wrong.
 Meta means about the thing itself, So Meta-Ethics is Ethics about Ethics.
Prescriptive Ethics (Normative Ethics)
Prescriptive Ethics, also known as, normative ethics can be defined as the study of ethical action,
typically based on what is morally right and wrong. Normative ethics is more practical as it
applies to basic human behavior and actions. There are mainly three theories that come under
normative ethics. Deontological, Teleological (Consequential), Virtue.
 Deals with questions like “is that action right (ethical)? OR was that act wrong?
 Checks if the action/outcome of action fits into the definition of right or wrong.

6. 2
 Includes Deontological Ethics (Focus on action/duty), Teleological Ethics (Focus on the
outcome/end), and Virtue Ethics are the sub-branches.
Prescriptive Ethics involves systematizing, defending, and recommending concepts of right and
wrong action (behavior). It also includes arriving at moral standards that guide to act right or
wrong. It is an ideal litmus test of identify behavior.
Descriptive Ethics (Comparative Ethics)
Descriptive ethics can be defined as describing and explaining people’s moral attitudes and the
moral norms and practices of societies. It deals with what is believed to be good, right, or
virtuous and moral practices societies do have.
Descriptive ethics are also known as comparative ethics and it is empirically based, and aim to
discover and describe the moral beliefs of a specific culture.
 Deals with people’s beliefs about morality.
 Deals with what society thinks is good or bad.
 It is an empirical investigation of the moral beliefs of various groups.
Applied Ethics
Applied ethics can be defined as a branch of moral philosophy that attempts to apply ethical
principles and moral theories to real-life moral issues. It is again divided into Bioethics, Cyber
Ethics, Environmental Ethics, Personal Ethics, Professional Ethics, Public Ethics, International
Ethics, and so on.
 The most practical branch of ethics.
 Deals with ethical questions specific to practical fields.
 Includes bioethics, legal ethics, business ethics, medical ethics, etc.
Ethics in Research
Research ethics are the set of ethical guidelines that guides us on how scientific research should
be conducted and disseminated. Research ethics govern the standards of conduct for scientific
researchers It is the guideline for responsibly conducting the research. Research ethics is
unambiguously concerned in the examination of ethical issues that are upraised when individuals
are involved as participants in the study [3].

7. 1
Objectives of research ethics
 To protect the rights, safety, and welfare of research participants.
 To promote the integrity, quality, and credibility of the research.
Principles of Research Ethics
The general principles of research ethics are [4]:
Table-1: Principles of Research Ethics
Honesty Being honest about the:
 Beneficiaries and respondents.
 Findings and methodology of the research.
 Direct and indirect stakeholders.
Integrity In order to keep the integrity b/n study subjects and researchers:
 Ensuring honesty and sincerity.
 Fulfilling agreements and your promises.
 Do not create false expectations or make false promises.
Objectivity Avoiding bias in experimental design, data analysis, data interpretation,
peer review, and other aspects of research.
Informed consent Is:
 A person knowingly, voluntarily and intelligently gives consent
to participate in a research.
 Related to the autonomous right of the individual to participate
in the research.
 Informing the participant about the research objective, their role,
benefits/harms (if any) etc.
Respect for
person/respondent
It includes:
 Autonomy, which requires that those who are capable of
deliberation about their personal goals should be treated with
respect for their capacity for self-determination; and

8. 2
 Protection of persons with impaired or diminished autonomy,
which requires that those who are dependent or vulnerable be
afforded security against harm or abuse.
Beneficence Maximize the benefits of the participants. Ethical obligation to maximize
possible benefits and to minimize possible harms to the respondents.
Non-maleficence/
Protecting the
subjects (human)
Do no harm. Minimize harm/s or risks to the human. Ensure privacy,
autonomy and dignity.
Responsible
publication
Responsibly publishing to promote and uptake research or knowledge.
No duplicate publication.
Protecting
anonymity
It means keeping the participant anonymous. It involves not revealing
the name, caste or any other information about the participants that may
reveal his/her identity.
Confidentiality Protecting confidential information, personnel records. It includes
information such as:
 Introduction and objective of the research
 Purpose of the discussion
 Procedure of the research
 Anticipated advantages, benefits/harm from the research (if any)
 Use of research
 Their role in research
 Right to refuse or withdraw
 Methods which will be used to protect anonymity and
confidentiality of the participant
 Freedom to not answer any question/withdraw from the research
 Who to contact if the participant needs additional information
about the research.
Non-discrimination Avoid discrimination on the basis of age, sex, race, ethnicity or other
factors that are violation of human rights and are not related to the study.
Openness Be open to sharing results, data and other resources. Also accept
encouraging comments and constructive feedback.

9. 3
Carefulness and
respect for
intellectual property
Be careful about the possible error and biases.
Give credit to the intellectual property of others. Always paraphrase
while referring to others article, writing. Never plagiarize.
Justice The obligation to distribute benefits and burdens fairly, to treat equals
equally, and to give reasons for differential treatment based on widely
accepted criteria for just ways to distribute benefits and burdens.
Broadly categorizing, there are mainly five principles of research ethics [3] [4]:
1. MINIMIZING THE RISK OF HARM
It is necessary to minimize any sort of harm to the participants. There are a number of forms of
harm that participants can be exposed to. They are:
 Bodily harm to contributors.
 Psychological agony and embarrassment.
 Social drawback.
 Violation of participant’s confidentiality and privacy.
In order to minimize the risk of harm, the researcher/data collector should:
 Obtain informed consent from participants.
 Protecting anonymity and confidentiality of participants.
 Avoiding misleading practices when planning research.
 Providing participants with the right to withdraw.
2. OBTAINING INFORMED CONSENT
One of the fundamentals of research ethics is the notion of informed consent. Informed consent
means that a person knowingly, voluntarily and intelligently gives consent to participate in a
research. Informed consent means that the participants should be well-informed about the:
 Introduction and objective of the research
 Purpose of the discussion
 Anticipated advantages, benefits/harm from the research (if any)
 Use of research
 Their role in research
 Methods which will be used to protect anonymity and confidentiality of the participant

10. 4
 Freedom to not answer any question/withdraw from the research
 Who to contact if the participant need additional information about the research
3. PROTECTING ANONYMITY AND CONFIDENTIALITY
Protecting the anonymity and confidentiality of research participants is an additionally applied
constituent of research ethics.
 Protecting anonymity: It means keeping the participant anonymous. It involves not
revealing the name, caste or any other information about the participants that may reveal
his/her identity.
 Maintaining confidentiality: It refers to ensuring that the information given by the
participant are confidential and not shared with anyone, except the research team. It is also
about keeping the information secretly from other people.
4. AVOIDING MISLEADING PRACTICES
 The researcher should avoid all the deceptive and misleading practices that might
misinform the respondent.
 It includes avoiding all the activities like communicating wrong messages, giving false
assurance, giving false information etc.
5. PROVIDING THE RIGHT TO WITHDRAW
Participants have to have the right to withdraw at any point of the research.
When any respondent decides on to withdraw from the research, they should not be stressed or
forced in any manner to try to discontinue them from withdrawing.
Apart from the above-mentioned ethics, other ethical aspects things that must be considered
while doing research are:
Protection of vulnerable groups of people:
 Vulnerability is one distinctive feature of people incapable to protect their moralities and
wellbeing. Vulnerable groups comprise captive populations (detainees, established,
students, etc.), mentally ill persons, and aged people, children, critically ill or dying, poor,
with learning incapacities, sedated or insensible.
 Their participation in research can be endorsed to their incapability to give an informed
consent and to the need for their further safety and sensitivity from the research/researcher
as they are in a greater risk of being betrayed, exposed or forced to participate.

11. 5
Principles of ethics in medical research
Principles of essentiality
Refers to whether the research is considered to be absolutely essential after a due consideration of
the existing scientific knowledge in the proposed area of research. This should be scrutinized by
an independent and responsible body of persons who, after careful consideration, come to the
conclusion that the research is likely to benefit the humanity or environment.
Principles of voluntariness, informed consent, and community agreement
Research participants should be fully apprised of the research and the associated risks and benefits.
The participants should be informed of the right to abstain from the research or withdraw consent
at any time. Where research entails treating any community, the principles of voluntariness and
informed consent apply to the community as a whole and to each individual member. In case a
person is incapable of giving consent, a legally acceptable guardian should give the informed
consent.
Principles of non-exploitation
The participants should be fully apprised of all the possible dangers that may arise during the
research so that they can appreciate all the physical and psychological risks. Each research should
include an in-built mechanism for compensation for the human participants either through
insurance cover or by any other appropriate means to cover foreseeable and unforeseeable risks,
and provide remedial action and comprehensive aftercare.
Principles of privacy and confidentiality
The identity and records of the participants are as far as possible kept confidential (except when
required for legal reasons). This is to avoid any form of hardship, discrimination or stigmatization
as a consequence of having participated in the research.
Principles of precaution and risk minimization
Due care and caution should be taken at all stages of the research and experiment to ensure that
the research participant and those affected by it including the community are put to the minimum
risk, suffer from no known irreversible adverse effects, and generally, benefit from the research or
experiment. There should be a plan for interim reviews to detect whether any intervention arm
(active or control) is associated with increased risks, so that undue harms are avoided by stopping
the research.

12. 6
Principles of professional competence
Research should be conducted by competent and qualified persons who act with total integrity and
impartiality and who have been made aware of the ethical considerations to be borne in mind in
respect of such research or experiment.
Principles of accountability and transparency
The research or experiment should be conducted in a fair, honest, impartial, and transparent
manner after full disclosure is made by those associated with the research or experiment of each
aspect of their interest in the research, and any conflict of interest that may exist. Full and complete
records of the research should be retained for such reasonable period as may be prescribed or
considered necessary for the purposes of post-research monitoring, evaluation of the research,
conducting further research, and scrutiny by the appropriate legal and administrative authority, if
necessary.
Principles of the maximization of the public interest and of distributive justice
The research or experiment and its subsequent application should be conducted and used to benefit
all human kind (and not just those who are socially better off), in particular, the research
participants themselves and or the community from which they are drawn.
Principles of public domain
The research findings should be brought into the public domain so that its results are generally
made known through scientific and other publications. This would help in consolidating the
scientific knowledge base of the field being studied and would prevent the undue replication of
studies which pose risks to some subjects.
Principles of totality of responsibility
Professional and moral responsibility should be observed, for the due observance of all the
principles, guidelines, or prescriptions of those directly or indirectly connected with the medical
research. This extends to the institutes where this research is carried out, as well as the sponsors
of the research. The research should be duly monitored and constantly subject to review and
remedial action at all stages [5].
Ethical Dilemma
An ethical dilemma (ethical paradox or moral dilemma) is a problem in the decision-making
process between two possible options, neither of which is absolutely acceptable from an ethical

13. 7
perspective. Although we face many ethical and moral problems in our lives, most of them come
with relatively straightforward solutions [6].
Ethical Dilemmas and Morals
Ethical dilemmas arise when moral issues raise questions that cannot be answered with a simple,
clearly defined rule, fact, or authoritative view.
Moral dilemmas occur when some evidence indicates that an act is morally right and some
evidence indicates the act is morally wrong; yet the evidence on both sides is inconclusive; or an
individual believes that on moral grounds, he or she cannot commit an act [7].
Approach to Solve an Ethical Dilemma
The following approaches to solve an ethical dilemma were deduced:
1. Refute the paradox (dilemma): The situation must be carefully analyzed. In some cases,
the existence of the dilemma can be logically refuted.
2. Value theory approach: Choose the alternative that offers the greater good or the lesser
evil.
3. Find alternative solutions: In some cases, the problem can be reconsidered, and new
alternative solutions may arise.
Research Misconduct
Research misconduct occurs when a researcher fabricates or falsifies data, or plagiarizes
information or ideas within a research process. Research misconduct means fabrication,
falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting
research results [8].
(a) Fabrication is making up data or results and recording or reporting them.
(b) Falsification is manipulating research materials, equipment, or processes, or changing or
omitting data or results such that the research is not accurately represented in the research
record.
(c) Plagiarism is the appropriation of another person's ideas, processes, results, or words
without giving appropriate credit.

14. 8
Ethical issues in my dissertation research
 Title: Early detection, classification and prediction of breast cancer using stacking
ensemble deep learning model.
 Objectives:
1. To develop a model that will used to diagnose and predict the likelihood of breast
cancer.
2. To integrate multiple deep learning models into a single new ensemble model to
improve the accuracy of predictions.
3. To assess the accuracy of breast cancer detection and prediction using deep learning
models.
Ethical principles in the use of deep learning in healthcare
Some of ethical principles in the use of deep learning in healthcare includes [9]:
 Transparency
 ML and AI system should be transparent, and their decision-making processes
should be explainable to users
 Human augmentation
 Designed to augment human capabilities, not replace them
 Bias evaluation
 Evaluated for bias and steps should be taken to address any bias that is identified
 Explainability by justification
 Explainable to users, and their decision-making processes should be transparent
 Algorithmic fairness and biases
 Designed to avoid bias and discrimination
 Promote fairness.
 Responsibility
 Responsible for their actions and the consequences of their decisions.
 Privacy
 Protect and hide their personal data

15. 1
Ethical dilemmas in deep learning
Some of the ethical dilemmas in deep learning includes [10]:
 Biased dataset
 When the algorithms are trained on biased datasets, the outcome becomes unfair
or discriminatory.
 Privacy
 It requires large amount of data, so raising concerns about privacy and data
protection.
 Transparency & Explainability
 Some algorithms can produce difficult results for humans to understand or explain
 Security
 Deep learning algorithms can be vulnerable to cyber attacks
Ethical dilemma in Breast cancer
Some of the ethical dilemma in Breast cancer includes [11]:
 Privacy concerns
 Collecting patient's medical information without their consent raises privacy
concerns.
 Access to care:
 Limited-resource in organization may face challenges in providing access to breast
cancer diagnosis and treatment
 False positives/negatives results
 If they receive a false negative result that delays necessary treatment.
Research Misconducts in Deep learning:
Some of the Research Misconducts in Deep learning [12]:
 Biased data collection
 Inappropriate use of personal information
 Lack of transparency
 Misrepresentation of results

16. 1
Research Misconducts in Breast cancer:
Some research misconducts in Breast cancer includes [13]:
 Falsification of data
 Researchers have been fixed falsifying data to make their studies appear more
significant.
 Fabrication of results
 Researchers have fabricated results entirely, causing harm to patients
 Failure to obtain informed consent
 Researchers have failed to obtain informed consent from patients,
Mitigation strategies for research dilemma and misconduct
Some of the mechanisms are [14]:
 Continuous education and training
 Institutional policies and procedures or standards
 Accountability
 Collaboration and communication
 Reporting and investigation
 Discussions on different research dilemma and misconduct

17. 2
Ethical Challenges and Mitigation strategies in my dissertation
Table-2: Ethical Challenges and Mitigation strategies
No Objectives Ethical challenges Mitigation strategies
I Data collection in
research
Lack of obtaining informed
consent from organization
Permission
Data security problem Implement strong data security
measures and adhere to privacy laws
and regulations
Biased data Deep learning algorithms is diverse
and representative of the population
to avoid bias in data.
II Training dataset
and Test set in deep
learning
Lack of transparency Retrain the dataset again and again with
new data to achieve transparent dataset
Biased Data Implementing systematic data
preprocessing techniques such as bias
detection and mitigation strategies used
to identify and reduce biases
Improper handling of personal
information
Implement data anonymization and
encryption techniques
III Research Method
(models) in deep
learning
Fairness selecting the model Evaluating performance metrics with
respect to fairness considerations

18. 3
Reference
1. https://www.canada.ca/en/what-is-ethics.html
2. https://edukedar.com/types-of-ethics
3. Kusum Wagle, 2020, Research ethics, https://www.publichealthnotes.com/research-ethics-
definition-principles-and-advantages/
4. http://dissertation.laerd.com/principles-of-research-ethics.php
5. Indian Council of Medical Research. Ethical Guidelines for Biomedical Research on Human
Subjects, New Delhi. 2006
6. CFI Team, 2020, https://corporatefinanceinstitute.com/resources/esg/ethical-dilemma/
7. Page K. The four principles: can they be measured and do they predict ethical decision making?
BMC Med Ethics. 2012 May 20;13:10. doi: 10.1186/1472-6939-13-10. PMID: 22606995;
PMCID: PMC3528420.
8. U.S. Department of Health and Human Services, Office of Research Integrity, Definition of
Research Misconduct | ORI - The Office of Research Integrity (hhs.gov)
9. https://ethical.institute/principles.html
10. Niral Sutaria, CISA, ACA, 2022, Bias and Ethical Concerns in Machine Learning,
https://www.isaca.org/resources/isaca-journal/issues/2022/volume-4/bias-and-ethical-
concerns-in-machine-learning.
11. Cummings NB. Ethical issues and the breast cancer patient. Arch Pathol Lab Med. 1994
Nov;118(11):1077-80. PMID: 7979890.
12. Juyang Weng, 2022, Research Misconducts in Deep learning
13. Gupta A. Fraud and misconduct in clinical research: A concern. Perspect Clin Res. 2013
Apr;4(2):144-7. doi: 10.4103/2229-3485.111800. PMID: 23833741; PMCID: PMC3700330.
14. Bornmann, Lutz. 2013. "Research Misconduct—Definitions, Manifestations and Extent"
Publications 1, no. 3: 87-98. https://doi.org/10.3390/publications1030087