Page 545
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue VI, June 2026
Consultancy Frameworks for AI-Driven Medical Imaging:
Advancing Early Diagnosis in Healthcare Systems
Hazel Galas Lampitoc*, Eduardo R. Yu II.
AMA University, Philippines.
DOI:
https://doi.org/10.51583/IJLTEMAS.2026.150600043
Received: 14 June 2026; Accepted: 18 June 2026; Published: 02 July 2026
ABSTRACT
While the integration of AI (artificial intelligence), its application in medical imaging is rapidly changing the
world of diagnostics, some health systems have yet to develop specific design guides to offer guidance for how
we make AI usage in medical imaging safe and ethically sound, whilst in a way that could scale to a wider
context with good security. We now present a consultancy system for AI-driven medical imaging for early
diagnosis in Saudi Arabia. Based on international standards like SPIRIT-AI, CONSORT-AI, CLAIM, DECIDE-
AI and the WHO Ethics and Governance of AI for Health, the framework covers essential obligations for
responsible AI use.
An expert perspective approach was used to gather the expertise in the field by utilizing qualitative design based
on healthcare professionals, radiology practitioners, biomedical engineering and digital health specialists
deployed in Saudi Arabia. Some of the more specific domains identified in this analysis relate to readiness
assessment, workflow integration, interoperability, data governance, cybersecurity, explainable AI (XAI),
federated learning, regulatory alignment with FDA, AI Act, and continuous metrics. The results indicate that the
successful introduction of AI in medical imaging should be achieved within an organisational consultancy model,
with ethical restrictions and technical validation, adaptation of clinical workflow and stakeholder participation.
The framework developed provides applicable recommendations for healthcare institutions working to
incorporate AI-assisted imaging tools to optimize early diagnosis, enhance efficiency in radiology, as well as to
promote global AI governance best practices. We add to the growing digital health literature by providing a
contextualized evidence-informed model aimed at supporting strategizing judgments and fast-tracking the move
to AI enabled diagnostic platforms in Saudi healthcare systems.
Keywords: Medical imaging AI-enabled; consultancy frameworks; early diagnosis tools; radiology workflow
integration; digital health transformation; explainable AI (XAI); federated learning; AI governance; SPIRIT-AI;
CONSORT-AI; CLAIM checklist; DECIDE-AI; FDA AI-enabled devices; EU AI Act; WHO AI ethics;
interoperability; data governance; biomedical engineering; Saudi healthcare settings.
INTRODUCTION
Background
Artificial Intelligence (AI) has played a key role in contemporary medical imaging, and machine learning has
begun to play a key role in image detection, triage, reporting and quality control in radiology. Findings from
large-scale trials suggest that AI systems can obtain clinically valuable performance if designed in strong data
sets and transparent validation (McKinney et al., 2020; Rajpurkar et al., 2022). Technical correctness does not
by itself bring clinical benefit. The practical use of AI in clinical practice will rely on how readily technology
can be integrated within radiology workflows, interoperability with legacy PACS/RIS systems, clinician
confidence, and the organization's systems for ongoing monitoring and governance (Tejani, Cook, et al., 2024;
Willemink et al., 2020). Many health care systems (including those in the process of digital transformation) are
focused on AI-mediated imaging in order to facilitate early diagnosis, expedited reporting of diagnosis and
standardised care flow. National strategies that underscore quality, efficiency, prevention, and digital health
innovation enable the promotion of good prospects for AI adoption (Alasiri & Mohammed, 2022). Yet adopting
Page 546
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue VI, June 2026
technology can’t be achieved on its own. Hospitals have to integrate AI deployment into clinical governance,
workflow design, data management, cybersecurity and ethical principles as proposed by international
organizations like the World Health Organization (2021).
Research Problem
Despite its application in early diagnosis and radiology, the adoption of AI in healthcare generally is patchy.
Adoption is often spearheaded by vendor-specific interventions, isolated pilot projects or department-level
experimentation. The key issue that this research tackles is that existing consultancy frameworks, which have
been validated in general, do not provide the necessary guidance for healthcare institutions to consider readiness
assessment, technology choice, workflow redesign, governance alignment, validation, deployment and constant
monitoring. Without such organization, AI adoption can be extremely heterogeneous among facilities, leading
to disparate results and the delayed arrival of diagnostic tools.
Research Gap
The existing literature offers reporting guidelines, governance guidance and technical standards for AI
applications in healthcare like CONSORT-AI, SPIRIT-AI, CLAIM, DECIDE-AI, and other evaluation guidance
frameworks (Cruz Rivera et al., 2020; Liu et al., 2020; Mongan et al., 2020; Vasey et al., 2022). Yet these
resources provide no consultancy-based model responsive to the operational realities of healthcare imaging
environments such as radiology workflows, PACS/RIS integration, early diagnosis and national digital health
strategies. This study seeks to fill this gap through developing and validating a contextualized consultancy model
for AI-driven medical imaging provision in healthcare systems on the basis of expert opinions from professionals
working in Saudi Arabia.
Objectives of the Study
Broadly, this study aims to create a consultancy framework for AI's adoption in medical imaging
implementations within healthcare systems in support of their successful execution and validation. In particular,
the study specifically seeks to:
Identify major barriers and enablers of adoption of AI in medical imaging environments.
Develop a consulting framework covering clinical, technical, organizational and governance aspects.
Evaluate the framework by experienced professionals on clarity, relevance, completeness, feasibility, and quality.
Iterate the framework after qualitative expert discussion and quantitative validation analyses.
Research Questions
The study thus is guided by the following questions:
1. What are the barriers and drivers to AI adoption in the context of medical imaging environment?
2. What core components should AI-driven early diagnosis be built on in terms of consultancy?
3. How do experts grade a proposed framework for clarity, relevance, completeness, practicality and
overall quality?
4. How can it be aligned with current AI governance standards and national digital health priorities?
Significance of the Study
This study has implications for clinical, technical, organizational, and governance audiences of health systems.
Clinically, it offers radiologists and imaging teams a stepwise pathway to implement AI without derailing clinical
Page 547
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue VI, June 2026
practice. Technically, it notes interoperability, data preparation, security and PACS/RIS integration as crucial
for AI adoption (Tejani, Cook, et al., 2024; Willemink et al., 2020). Strategically, it connects AI adoption to
more extensive priorities of health transformation. At the governance level, it materializes principles of
transparency, accountability, human oversight, safety, privacy, and continuous reassessment in actionable steps
(Amann et al., 2020; European Parliament and Council of the European Union, 2024; World Health Organization,
2021).
LITERATURE REVIEW
Medical Imaging, early diagnosis in AI systems.
AI-based imaging systems have been widely explored in a variety of aspects such as cancer screening, diagnostic
support, triage, segmentation, and clinical decision support. For AI-based systems, this has led researchers to
develop innovative methods for early diagnosis, to detect disease in the initial stages of illness as well as for
early diagnosis. Large-scale evaluations such as that from McKinney et al., for example, the international
assessment of an AI system for breast cancer screening (2020), have indicated that AI is capable of clinically
meaningful performance in particular image scanning tasks. General summaries of AI in health and medicine
have shown that clinical benefit is contingent not only on the accuracy of algorithms but also on safe design of
programs, a representative model and practice that uses representative datasets which could be evaluated in the
future, and responsible deployment (Rajpurkar et al.). Not only do these results indicate AI's prospects for
enhancing early diagnosis, but they make clear that technical performance is not enough for successful clinical
integration on its own. This paper underscores the fact that data preparation is key to a successful integration of
AI technology in medical imaging. Willemink et al. (2020) emphasize curated, annotated, standardized and well-
governed imaging datasets for machine learning results. Roberts et al. (2021) also warn that poor data sources,
inconsistent validation of results, or inadequate reporting can compromise imaging AI research reliability. This
knowledge can validate the strategic incorporation of data pre-processing, validation, and continuous monitoring
in the consultancy model presented in this study.
Readiness, acceptance, and human aspects.
The adoption of AI in healthcare is a sociotechnical process and necessitates harmonizing people, process,
systems and institutional readiness. Human-centric assessments suggest that even if AI tools perform well, there
may still be adoption barriers, such as lack of fit with clinical experience, unmet user requirements, or operational
limitations (Abràmoff et al., 2020; Chen et al., 2021). In the Saudi context, national digital health programmes
following Vision 2030 have stimulated interest in AI-enabled imaging but its adoption remains an issue of
workforce readiness, organisational maturity, governance frameworks and clinician confidence (Alasiri &
Mohammed, 2022; Alsaedi et al., 2024). Trust in AI is also strongly attached to explainability and accountability.
Amann et al. (2020) maintain that explainability in healthcare AI must be understood from clinical, ethical,
technical, and social vantage points. According to the study conducted by Tjoa and Guan (2021) and by van der
Velden et al. (2022) demonstrates that explainable AI (XAI) can facilitate transparency with respect to medical
image analysis, while methods of explanation must be interpreted with care. These results underscore the
importance of integrating governance, clinical acceptability, and human centered design into AI development
frameworks.
Workflow integration, interoperability and technical governance.
A successful deployment of AI in radiology necessitates an integration of the technology with current imaging
workflow such as image acquisition, worklist management, PACS/RIS routing, reporting, communication of
results, and quality assurance. Tejani, Cook, et al. (2024) highlight that interoperability through standards-based
frameworks is an absolute must for scaling AI in radiology. The issue is not simply whether an AI model can
generate an output, but whether that output reaches the right person at the right time in a hospital’s clinical
workflow. Commercial AI products, on the other hand, differ widely in the extent of scientific evidence to
support clinical use. van Leeuwen et al. (2021) discovered that a number of radiology AI products differ greatly
in their supporting evidence quality and extent, and as a result structured evidence review is crucial prior to
purchase or installation. Privacy preserving, federated learning solutions are becoming more relevant, notably
Page 548
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue VI, June 2026
when healthcare facilities cannot easily collect sensitive imaging information. Rieke et al. (2020), Kaissis et al.
(2020), and Sharma and Guleria (2023, respectively) identify federated learning as one exciting avenue for multi-
institutional AI development—but with challenges of security, generalizability, governance, and technical
complexity.
Governance, Ethics, and Regulatory expectations.
In response to this, responsible AI implementation necessitates the implementation of governance structures
covering safety, accountability, transparency, bias, privacy, as well as post-deployment monitoring, especially
in a way that makes it possible for individuals to recognize and respond to the risks as they arise. International
guidelines from the World Health Organization (2021) and the European Union Artificial Intelligence Act
(European Parliament and Council of the European Union, 2024) highlight principles including human oversight,
fairness, privacy, security and accountability. Reddy et al. (2020) and Morley et al. (2020), governance needs
processes to operationalize, through policies, rules regarding roles, risk mitigation mechanisms, and ethics
reviews, rather than simply at the level of the broader principles. Regulations related to AI-enabled medical
devices are ever-evolving. Benjamens et al. (2020) reported the rapid proliferation of AI/ML–based medical
devices that show regulatory authorization, and the U.S. Food and Drug Administration (n.d.) has a public list
of AI–enabled medical devices it publishes. These developments show healthcare systems should examine
vendor claims, regulatory standing, validation evidence, and clinical risk carefully before introducing AI
solutions. There are numerous reporting and evaluation guidelines regarding the development and assessment of
AI for medical imaging. The Checklist for Artificial Intelligence in Medical Imaging (CLAIM) offers a
structured and uniform reporting regime for authors and reviewers (Mongan et al., 2020), with the 2024 update
reflecting evolving expectations for AI reporting (Tejani, Klontzas et al., 2024). CONSORT–AI and SPIRIT–
AI advance clinical trial reporting and protocol standards for AI-based interventions (Cruz Rivera et al., 2020;
Liu et al., 2020). DECIDE–AI provides a foundation framework for early-stage clinical evaluation of AI-driven
decision-support systems (Vasey et al., 2022). Vollmer et al. (2020) also propose critical questions in relation to
transparency, reproducibility, ethics, and effectiveness that must be solved in medical machine learning research.
These standards also influence the consultancy framework that focuses on validation, transparency, stakeholder
involvement and responsible reporting. 2.6 Synthesis. The literature indicates that there is great promise in AI
medical imaging based system in advancing early diagnosis, but successful AI in medical imaging
implementation should be approached with a degree of caution. Safe and successful adoption demands
organisational preparedness, integration of workflow, compliance, explainability, ethical governance, clinical
evaluation and ongoing assessment. Despite the fact that existing guidances cover much of these areas, they do
not involve a consultancy-based solution customized to the operational aspects of healthcare imaging settings.
This study aims to address this void by developing a systematic consulting framework that combines strategic
alignment, clinical workflow, technical infrastructure, and governance for one unified AI-based medical imaging
adoption model, for medical imaging.
METHODS
Research Design.
This study also followed a sequential mixed-methods design in developing and validating a consultancy
framework that helped AI-driven medical imaging implementation in healthcare systems. The qualitative phase
sought out expert perspectives on barriers to adoption, workflow needs, technical integration, and governance
considerations. The quantitative element involved a structured validation questionnaire to evaluate proposed
structure. The nature of the design proposed was justified by the study's need for contextual and measurable
expert examination.
Framework Design Process.
The framework was formulated in six phases: problem analysis, framework design, prototype construction,
expert testing, data analysis, and refinement. These were consistent with the existing literature on AI reporting,
governance and clinical assessment (Liu et al., 2020; Mongan et al., 2020; Vasey et al., 2022).
Page 549
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue VI, June 2026
Stage 1: Determine adoption barriers or implementation needs in AI-assisted imaging.
Stage 2: Develop a model based on the integrated structure with clinical and technical, organisational,
as well as governance elements.
Stage 3: Development of a validation instrument and framework prototype.
Stage 4: Carry out expert review: Interviews and Survey evaluation.
Stage 5: Qualitative and quantitative findings.
Stage 6: Consultation structure and experts feedback to perfect the consultancy architecture.
Participants and Sampling.
It was a purposive sampling study to get experts that are experts on radiology, biomed engineering and
PACS/RIS administration; digital health, AI governance, AI imaging deployment. The evaluators and
respondents included healthcare practitioners who currently work in hospitals and medical establishments in
Saudi Arabia. We did not collect any institutional data from the Ministry of Health, nor did we need an
authorization because we focused primarily on expert assessment rather than organizational record. The number
of experts was 17 in total. This sample size provided an appropriate sample size for exploratory framework
validation, as the goal is expert-informed assessment rather than statistical generalization.
Table 1.
Participant Group
Role in AI-Driven Medical Imaging
Number
Radiologists
Assess clinical usefulness, diagnostic workflow, and early
diagnosis value
4
Radiology department heads
Assess operational fit, workflow redesign, and departmental
feasibility
2
Biomedical engineers
Assess equipment, safety, and technical feasibility
3
PACS/RIS administrators
Assess data flow, system integration, interoperability, and
reporting linkage
2
Digital health & AI governance
officers
Assess policy alignment, governance, cybersecurity, and
compliance
4
AI imaging vendor specialists
Assess technical deployment requirements and product
behavior
2
Total
17
Note. Participant groups reflect the 17 expert validators recruited through purposive sampling.
Criteria for Inclusion and Exclusion
Participants were required to have at least 3 years of experience in the areas of radiology, biomedical engineering,
digital health, AI governance, PACS/RIS systems, or medical imaging technology as well as an ability to critique
the suggested method of analysis. Exclusions were irrelevant experience, no participation in AI decision-making
or implementation, and unwillingness to complete the interview or questionnaire.
Research Instruments
Page 550
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue VI, June 2026
Two instruments were used:
1.Semi-structured interview guide – examined barriers, issues in functioning/workflows, technical
integration, governance, consultancy,
2. Anticipated framework elements item Likert-scale questionnaire – assessed the clarity of framework,
relevance, completeness, practicality, and overall evaluation.
14-Responses ranged from 1 (strongly disagree) to 5 (strongly agree).
Table 2. Surveys for Framework Validation
Construct
Code
Survey Item
Clarity
C1
The framework is clearly structured.
C2
The components are easy to understand.
C3
The workflow is logical and coherent.
Relevance
R1
The framework addresses real challenges in AI imaging.
R2
The framework is relevant to radiology workflows in healthcare settings. (MOH
reference removed)
R3
The framework aligns with clinical and technical needs.
Completeness
CP1
The framework includes the necessary implementation components.
CP2
The consultancy processes are comprehensive.
CP3
Governance and policy considerations are adequately covered.
Practicality
P1
The framework can be realistically implemented.
P2
The framework supports early diagnosis using AI imaging.
P3
The framework improves decision-making for AI adoption.
Overall
Evaluation
O1
I am satisfied with the framework.
O2
I would recommend the framework for adoption in healthcare systems.
Data Collection Procedure
Data was collected in two phases. In the qualitative phase, researchers examined the original framework and
conducted semi-structured interviews to probe about barriers to adoption, workflow fit, technical integration,
governance, and consultancy needs. During the quantitative phase, the same specialists who were employed as
the initial team members filled out the 14-item validation questionnaire. This progressive approach allowed
qualitative views to inform the framework and quantitative results to assess its perceived quality and reliability.
Data Analysis
Qualitative data were analyzed with thematic analysis: familiarization, initial coding, theme development,
review, and interpretation. Topics included alignment with clinical workflow, technical integration and
interoperability, governance and regulatory compliance, readiness of the organization, and consultancy.
Descriptive statistics and reliability testing were used for quantitative data analysis. Means, standard deviations,
and score ranges were calculated for each item. Composite scores were computed by averaging items under each
Page 551
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue VI, June 2026
construct. Cronbach’s alpha was used to assess internal consistency. Because the sample was purposively chosen,
findings were reported descriptively rather than statistically; the findings were reframed for use not as
population-level generalizations.
Ethical Considerations
The study did not include patient records, clinical images, or identifiable patient names. Participation was
voluntary and informed consent was obtained prior to data collection. Responses from interviewees were
anonymized, and study data were securely stored. Finally, any ethics approval number given by the institution
was included in the journal submission.
RESULTS
Overview of Findings
Results connect qualitative themes as well as quantitative validation scores. In general, the consultation
framework was deemed clear, relevant, comprehensive, and practical as a reference plan for formal AI-based
medical imaging adoption in health systems. These results all signal a strong expert agreement on the benefits
of the framework in terms of the early diagnosis and responsible adoption of AI in healthcare systems.
Qualitative Themes
We made a thematic analysis to identify 5 central themes and subthemes. The themes are presented in Table 3
along with expert interview insights and findings and their relevance to the consultancy framework. These
insights reveal how expert views were translated into applicable actionable guidelines for AI-led imaging
adoption.
Table 3. Qualitative Themes, Subthemes and Implications for the Framework.
Theme
Subthemes
Framework Implication
Clinical workflow
alignment
Early diagnosis pathways;
radiologist workload; reporting
integration
AI tools should support clinical
decision-making without disrupting routine
radiology practice.
Technical integration
and interoperability
PACS/RIS compatibility; metadata
standards; distributed retrieval
Deployment must address data flow,
interoperability, and technical reliability.
Governance and
regulatory compliance
Regulatory requirements; data
privacy; cybersecurity; AI ethics
Implementation should incorporate
accountable governance and risk controls.
Organizational readiness
Infrastructure maturity; workforce
skills; change management
Healthcare facilities require readiness
assessment and capacity-building before
deployment.
Consultancy
requirements
Readiness assessment;
implementation roadmap;
evaluation metrics
A structured consultancy process can reduce
fragmented or inconsistent adoption.
Note. Themes were derived from semi-structured interviews with expert validators.
Descriptive Statistics
Descriptive statistics of the 14 validation items are shown in Table 4. Expert consensus for the quality of the
proposed work presented on the framework was robust, with means of all items over 4.29.
Page 552
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue VI, June 2026
Item
Mean
SD
Minimum
Maximum
C1
4.47
0.62
3
5
C2
4.41
0.71
3
5
C3
4.53
0.62
3
5
R1
4.35
0.70
3
5
R2
4.47
0.62
3
5
R3
4.41
0.62
3
5
CP1
4.35
0.70
3
5
CP2
4.29
0.77
3
5
CP3
4.41
0.62
3
5
P1
4.47
0.62
3
5
P2
4.53
0.62
3
5
P3
4.41
0.62
3
5
O1
4.47
0.62
3
5
O2
4.53
0.62
3
5
Note. N = 17. Responses were rated from 1 = strongly disagree to 5 = strongly agree.
Table 4. Summary statistics for the validation items
Composite Scores at the Construct-Level. Composite means were calculated as the average values of the items
under each construct. All constructs were scored extremely highly showing expert support of the framework.
Table 5. Composite Scores for Framework Evaluation Constructs
Construct
Composite Mean
Interpretation
Clarity
4.47
Very high
Relevance
4.41
Very high
Completeness
4.35
Very high
Practicality
4.47
Very high
Overall evaluation
4.50
Very high
Page 553
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue VI, June 2026
Note. Composite means were computed by averaging the items under each construct.
Figure 1 provides a visual summary of the composite mean scores.
Reliability Analysis.
The Cronbach’s alpha values varied from 0.81 to 0.92, indicating strong to excellent internal consistency across
all constructs.
Table 6. Results of Cronbach’s Alpha Reliability.
Construct
Cronbach’s Alpha
Interpretation
Clarity
0.84
Strong reliability
Relevance
0.89
Strong reliability
Completeness
0.81
Strong reliability
Practicality
0.86
Strong reliability
Overall evaluation
0.92
Excellent reliability
Note. Cronbach’s alpha values above 0.70 indicate acceptable internal consistency.
Expert Consensus Sum.
The qualitative and quantitative results were similar, indicating strong expert consensus. Experts stressed the
need for an organized governance-aware, workflow-sensitive implementation model for AI-driven medical
imaging. They rated the framework on their survey to confirm clarity, relevance, completeness, and practicality.
While the purposive sample does not imply population-level generalization, the findings offer a credible basis
for the platform to be honed and pilot tested by the healthcare ecosystem—especially that of healthcare
organizations which are undergoing digital transformation and focused on expanding early diagnosis via AI-
enabled imaging.
DISCUSSION
Interpretation of the Key Results.
The results indicate substantial expert support for the consultancy model. The high clarity, practicality, and
overall evaluation reviews suggest that for experts, the framework was perceived as understandable,
operationally applicable, and useful to steer the adoption of AI medical imaging. The qualitative themes support
Page 554
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue VI, June 2026
that AI adoption in imaging should be viewed as a cohesive institutional process rather than merely as the
acquisition of new technology. This reading is consistent with the literature which reiterates that successful
adoption of healthcare AI depends, firstly, on clinical validation and second, on organizational readiness,
governance structures, and integration into workflow (Abràmoff et al., 2020; Reddy et al., 2020; Vollmer et al.,
2020).
Fit with Medical Imaging AI Literature.
The focus on workflow integration and interoperability aligns well with modern research in radiology AI. Tejani,
Cook, et al. (2024) emphasise that standards-based interoperability underpins scalable AI deployment, whereas
Willemink et al. (2020) highlight the significance of data prep, annotation quality, and standardization of imaging.
PACS/RIS compatibility, metadata standards, and distributed retrieval were similarly found by the experts in
this study to be critical implementation issues. These results validate the application of the framework rather
than acting as a generic AI adoption checklist; addressing concrete clinical needs and technical imperatives.
Governance and AI Trustworthiness implications.
The governance aspect of the framework is reinforced by the international guidance on trustworthy AI. Principles
of transparency, explainability, human oversight, privacy, accountability, and safety are echoed in the world's
standards including the WHO guidance on AI ethics (World Health Organization, 2021) and the European
Union’s Artificial Intelligence Act (European Parliament and Council of the European Union, 2024). These
principles are operationalized through readiness assessment, regulatory screening and risk evaluation, validation
evidence, post-deployment monitoring and role-based accountability. This establishes an explicit bridge between
the ethical principles and applied implementation paths.
Strategic Implications for Digital Transformation Initiatives in Healthcare System Transformation.
The framework is strategically sensitive for healthcare systems, including Saudi Arabia, that are in the midst of
digital transformation and value quality improvement, prevention, access, and efficiency priorities (Alasiri &
Mohammed, 2022). AI-led imaging can help reach these objectives by being developed as a systematic,
evidence-based and clinically accepted process. The expert validation results imply that the framework may aid
health care organizations in assessing readiness, preventing piecemeal adherence, and aligning AI application
with national digital health goals without the need for institutional consent or access to organization records.
Clinical and Operational Implications.
Clinically, the framework helps early diagnosis by providing guideline for healthcare institutions to choose high-
value, evidence-supported and workflow-compatible AI use cases. At an operational level, it defines the roles of
radiologists, biomedical engineers, PACS/RIS administrators, digital health officers, governance personnel, and
vendors. It also encourages phased implementation—from readiness assessment to pilot testing and continued
monitoring. This is consistent with reliable reporting principles (Cruz Rivera et al., 2020; Liu et al., 2020;
Mongan et al., 2020; Vasey et al., 2022) and evaluation criteria CLAIM, CONSORT-AI, SPIRIT-AI, and
DECIDE-AI.
Limitations. Several limitations can be found within this study.
One, the validation involved 17 experts; this would be adequate for exploratory framework development but not
statistical generalization. Second, the study gauged expert perceptions, not real-world clinical outcomes. Third,
no imaging data at patient level was analyzed. Fourth, the proposed framework has not been empirically tested
with a live implementation pilot among healthcare institutions. Lastly, regulatory and digital health standards
change over time, and the framework needs to periodically be updated to ensure its continued relevance.
Page 555
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue VI, June 2026
Analysis Proposed Consultancy Framework
Supported by a synthesis of literature review, qualitative themes, and quantitative validation findings, the revised
consultancy framework is structured into seven planned implementation phases. Each stage, relevant consultancy
stage and possible outputs are summarised in Table 7.
Stage
Consultancy Activity
Expected Output
1. Readiness assessment
Assess clinical need, infrastructure, data
availability, workforce readiness, governance
maturity, and leadership support.
Readiness report and gap
analysis
2. Use-case prioritization
Identify high-value imaging use cases linked to
early diagnosis, patient safety, workload reduction,
or service quality.
Prioritized AI imaging
use-case list
3. Evidence and
regulatory review
Review validation evidence, regulatory status,
vendor documentation, bias risks, and
explainability claims.
Evidence and compliance
review
4. Data and system
integration planning
Map data flow, PACS/RIS integration, DICOM
routing, metadata requirements, cybersecurity,
privacy, and fallback procedures.
Technical integration plan
5. Clinical validation and
pilot deployment
Test the AI tool in a controlled workflow with
human oversight, performance monitoring, and
user feedback.
Pilot evaluation report
6. Full implementation
and change management
Train users, define roles, finalize governance
controls, and integrate AI outputs into routine
reporting pathways.
Implementation plan and
training record
7. Monitoring and
continuous improvement
Track performance, safety incidents, bias,
workflow impact, user acceptance, and model
drift.
Post-deployment monitoring
dashboard and review cycle
Note. The framework is intended as an implementation guide for healthcare systems and should be adapted based
on local readiness, regulatory context, and policy requirements.
The architecture offers a well-defined step by step pathway for healthcare organizations leading from readiness
assessment to use-case selection, integration planning, pilot validation, full deployment, and continued
monitoring. The framework is designed as an implementation guide for healthcare systems, and it would likely
be adapted to reflect specific local readiness, regulatory context and policy needs. This conceptualization places
consultancy as an evidence-informed, governance-aligned, structured process for safe, explainable, and
sustainable AI uptake of medical imaging. It is intended to be applicable to health care institutions of all stages
of digital maturity, and particularly to ones going through national digital transformation programs.
CONCLUSION AND IMPLICATIONS
Conclusion
This study created and validated a consultancy framework for AI-driven medical imaging implementation in
healthcare systems. Expert evaluation performed showed very high ratings for clarity, relevance, completeness,
practicality, and overall quality, which were supported by high reliability across all constructs. Qualitative results
validated that workflow alignment, PACS/RIS interoperability, governance compliance, organizational
readiness, and structured consultancy support are critical for successful adoption of AI in radiology. Using the
Page 556
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue VI, June 2026
literature reviewed, this framework incorporates clinical, technical, organizational, and governance activities as
a staged implementation pathway. It provides an evidence-informed, structured way for healthcare institutions
to adopt AI-enabled imaging tools without compromising their credibility and in a safe and accountable way.
The framework also assists with diagnosing early and increasing workflow efficiency while contributing to
broader digital health transformation objectives, making it adaptable to healthcare contexts where levels of
digital maturity may differ.
Recommendations
From the findings above, the following recommendations are presented for healthcare organizations planning
the implementation of AI-infused medical imaging. Trial the consultancy framework in healthcare organizations
with different levels of digital maturity to assess adaptability and operational feasibility. Establish radiology-
specific indicators of AI readiness to guide departments in the path of AI deployment. Set up a governance
checklist based on AI ethics guidelines for national, data protection, and health services regulatory requirements.
Require detailed vendor evidence reviews prior to procurement, with external validation, regulatory status,
integration, cybersecurity, and monitoring plan requirements. Assess post-deployment metrics including
reporting turnaround time, diagnostic precision, radiologists workload, user acceptance, and patient safety
metrics. Look into privacy preserving implementation models—like federated learning—when building a
multiple-site model or distributed data environments are necessary, e.g. for multi-site model development or
distributed environments for data processing.
Future Research
Future research needs to evaluate the framework through real world pilot implementations in healthcare facilities
before and after the adoption and compare results between pre- and post-adoption. Further research may explore
how the framework can apply to other radiology subspecialties, assess cost-effectiveness, and assess long-term
clinical and operational implications. Insights on governance models, regulatory evolution and sustainable
lifecycle management of AI imaging systems will strengthen the evidence base for responsible AI deployment
in healthcare.
Declarations
ACKNOWLEDGMENT
The authors appreciate the expert validators who have shared their professional expertise in radiology
workflow, biomedical engineering, PACS/RIS administration, digital health, AI governance, and medical
imaging implementation with us. Their expert knowledge was important for the validation and adjustment of
the proposed consultancy model.
Author Contributions
Engr. Hazel Galas Lampitoc led conceptualization, framework development, data collection, analysis, and
manuscript drafting. Dr. Eduardo R. Yu II supervised, conducted a methodology review, recommended critical
revisions, and reviewed and approved the manuscript. Both authors reviewed and approved the final manuscript.
Funding. This study did not receive external funding.
Conflict of Interest
The authors declare no conflict of interest. Data Availability. On reasonable request, summary data can be made
available anonymously by the corresponding author. The confidentiality limitation prevents dissemination of
raw interview transcripts and identifiable expert responses.
Ethics consent and informed consent
The study included expert-opinion validation only and no documentation of patient records, or images of patient
clinical problems, or identification of patient health information. Hence, it was classified as minimal-risk or
Page 557
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue VI, June 2026
exempt research. Participation was voluntary and informed consent was obtained before data collection.
Anonymisation of interview data and secure storage of data were maintained for all interview transcripts. They
should remain with the institution or journal for institutional confirmation of exemption as needed.
AI Assistance Disclosure
AI-assisted editing was employed only for grammar, clarity, format, and reference alignment. All responsibility
for study design, data collection, analysis, interpretation and final manuscript content is still attributable to
authors.
REFERENCES
1. Abràmoff, M. D., Tobey, D., & Char, D. S. (2020). Lessons learned about autonomous AI: Finding a
safe, efficacious, and ethical path through the development process. American Journal of
Ophthalmology, 214, 134–142. https://doi.org/10.1016/j.ajo.2020.02.022
2. Alasiri, A. A., & Mohammed, V. (2022). Healthcare transformation in Saudi Arabia: An overview
since the launch of Vision 2030. Health Services Insights, 15, 11786329221121214.
https://doi.org/10.1177/11786329221121214
3. Alsaedi, A. R., Alneami, N., Almajnoni, F., Alamri, O., Aljohni, K., Alrwaily, M. K., Eid, M., Budayr,
A., Alrehaili, M. A., Alghamdi, M. M., & Eid, M. H. (2024). Perceived worries in the adoption of
artificial intelligence among healthcare professionals in Saudi Arabia: A cross-sectional survey study.
Nursing Reports, 14(4), 3706–3721. https://doi.org/10.3390/nursrep14040271
4. Amann, J., Blasimme, A., Vayena, E., Frey, D., & Madai, V. I. (2020). Explainability for artificial
intelligence in healthcare: A multidisciplinary perspective. BMC Medical Informatics and Decision
Making, 20, Article 310. https://doi.org/10.1186/s12911-020-01332-6
5. Benjamens, S., Dhunnoo, P., & Meskó, B. (2020). The state of artificial intelligence-based FDA-
approved medical devices and algorithms: An online database. npj Digital Medicine, 3, Article 118.
https://doi.org/10.1038/s41746-020-00324-0
6. Chen, I. Y., Pierson, E., Rose, S., Joshi, S., Ferryman, K., & Ghassemi, M. (2021). Ethical machine
learning in healthcare. Annual Review of Biomedical Data Science, 4, 123–144.
https://doi.org/10.1146/annurev-biodatasci-092820-114757
7. Cruz Rivera, S., Liu, X., Chan, A.-W., Denniston, A. K., Calvert, M. J., Ashrafian, H., Beam, A. L.,
Collins, G. S., Darzi, A., Deeks, J. J., ElZarrad, M. K., Espinoza, C., Esteva, A., Faes, L., Ferrante di
Ruffano, L., Fletcher, J., Golub, R., Harvey, H., Haug, C., ... Yau, C. (2020). Guidelines for clinical
trial protocols for interventions involving artificial intelligence: The SPIRIT-AI extension. The Lancet
Digital Health, 2(10), e549–e560. https://doi.org/10.1016/S2589-7500(20)30219-3
8. European Parliament and Council of the European Union. (2024). Regulation (EU) 2024/1689 of 13
June 2024 laying down harmonised rules on artificial intelligence (Artificial Intelligence Act). Official
Journal of the European Union. https://eur-lex.europa.eu/eli/reg/2024/1689/oj/eng
9. Food and Drug Administration. (n.d.). Artificial intelligence-enabled medical devices. Retrieved June
10, 2026, from https://www.fda.gov/medical-devices/software-medical-device-samd/artificial-
intelligence-enabled-medical-devices
10. Kaissis, G. A., Makowski, M. R., Rückert, D., & Braren, R. F. (2020). Secure, privacy-preserving and
federated machine learning in medical imaging. Nature Machine Intelligence, 2(6), 305–311.
https://doi.org/10.1038/s42256-020-0186-1
11. Liu, X., Cruz Rivera, S., Moher, D., Calvert, M. J., Denniston, A. K., & the SPIRIT-AI and
CONSORT-AI Working Group. (2020). Reporting guidelines for clinical trial reports for interventions
involving artificial intelligence: The CONSORT-AI extension. The Lancet Digital Health, 2(10),
e537–e548. https://doi.org/10.1016/S2589-7500(20)30218-1
12. McKinney, S. M., Sieniek, M., Godbole, V., Godwin, J., Antropova, N., Ashrafian, H., Back, T.,
Chesus, M., Corrado, G. S., Darzi, A., Etemadi, M., Garcia-Vicente, F., Gilbert, F. J., Halling-Brown,
M., Hassabis, D., Jansen, S., Karthikesalingam, A., Kelly, C. J., King, D., ... Shetty, S. (2020).
International evaluation of an AI system for breast cancer screening. Nature, 577(7788), 89–94.
https://doi.org/10.1038/s41586-019-1799-6
Page 558
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue VI, June 2026
13. Ministry of Health. (2024, April 30). HSTP guide.
https://www.moh.gov.sa/en/Ministry/vro/Pages/manual.aspx
14. Mongan, J., Moy, L., & Kahn, C. E., Jr. (2020). Checklist for artificial intelligence in medical imaging
(CLAIM): A guide for authors and reviewers. Radiology: Artificial Intelligence, 2(2), Article e200029.
https://doi.org/10.1148/ryai.2020200029
15. Morley, J., Machado, C. C. V., Burr, C., Cowls, J., Joshi, I., Taddeo, M., & Floridi, L. (2020). The
ethics of AI in health care: A mapping review. Social Science & Medicine, 260, Article 113172.
https://doi.org/10.1016/j.socscimed.2020.113172
16. Rajpurkar, P., Chen, E., Banerjee, O., & Topol, E. J. (2022). AI in health and medicine. Nature
Medicine, 28(1), 31–38. https://doi.org/10.1038/s41591-021-01614-0
17. Reddy, S., Allan, S., Coghlan, S., & Cooper, P. (2020). A governance model for the application of AI
in health care. Journal of the American Medical Informatics Association, 27(3), 491–497.
https://doi.org/10.1093/jamia/ocz192
18. Rieke, N., Hancox, J., Li, W., Milletari, F., Roth, H. R., Albarqouni, S., Bakas, S., Galtier, M. N.,
Landman, B. A., Maier-Hein, K., Ourselin, S., Sheller, M., Summers, R. M., Trask, A., Xu, D., Baust,
M., & Cardoso, M. J. (2020). The future of digital health with federated learning. npj Digital Medicine,
3, Article 119. https://doi.org/10.1038/s41746-020-00323-1
19. Roberts, M., Driggs, D., Thorpe, M., Gilbey, J., Yeung, M., Ursprung, S., Aviles-Rivero, A. I., Etmann,
C., McCague, C., Beer, L., Weir-McCall, J. R., Teng, Z., Gkrania-Klotsas, E., Ruggiero, A., Korhonen,
A., Jefferson, E., Ako, E., Langs, G., Gozaliasl, G., ... Schönlieb, C.-B. (2021). Common pitfalls and
recommendations for using machine learning to detect and prognosticate for COVID-19 using chest
radiographs and CT scans. Nature Machine Intelligence, 3, 199–217. https://doi.org/10.1038/s42256-
021-00307-0
20. Sharma, S., & Guleria, K. (2023). A comprehensive review on federated learning-based models for
healthcare applications. Artificial Intelligence in Medicine, 146, Article 102691.
https://doi.org/10.1016/j.artmed.2023.102691
21. Tejani, A. S., Cook, T. S., Hussain, M., Sippel Schmidt, T., & O'Donnell, K. P. (2024). Integrating
and adopting AI in the radiology workflow: A primer for standards and Integrating the Healthcare
Enterprise (IHE) profiles. Radiology, 311(3), Article e232653. https://doi.org/10.1148/radiol.232653
22. Tejani, A. S., Klontzas, M. E., Gatti, A. A., Mongan, J., Moy, L., Park, S., & Kahn, C. E., Jr. (2024).
Checklist for artificial intelligence in medical imaging (CLAIM): 2024 update. Radiology: Artificial
Intelligence, 6(4), Article e240300. https://doi.org/10.1148/ryai.240300
23. Tjoa, E., & Guan, C. (2021). A survey on explainable artificial intelligence (XAI): Toward medical
XAI. IEEE Transactions on Neural Networks and Learning Systems, 32(11), 4793–4813.
https://doi.org/10.1109/TNNLS.2020.3027314
24. van der Velden, B. H. M., Kuijf, H. J., Gilhuijs, K. G. A., & Viergever, M. A. (2022). Explainable
artificial intelligence (XAI) in deep learning-based medical image analysis. Medical Image Analysis,
79, Article 102470. https://doi.org/10.1016/j.media.2022.102470
25. van Leeuwen, K. G., Schalekamp, S., Rutten, M. J. C. M., van Ginneken, B., & de Rooij, M. (2021).
Artificial intelligence in radiology: 100 commercially available products and their scientific evidence.
European Radiology, 31, 3797–3804. https://doi.org/10.1007/s00330-021-07892-z
26. Vasey, B., Nagendran, M., Campbell, B., Clifton, D. A., Collins, G. S., Denaxas, S., Denniston, A. K.,
Faes, L., Geerts, B., Ibrahim, M., Liu, X., Mateen, B. A., Mathur, P., McCradden, M. D., Morgan, L.,
Ordish, J., Rogers, C., Saria, S., Ting, D. S. W., ... McCulloch, P. (2022). Reporting guideline for the
early-stage clinical evaluation of decision support systems driven by artificial intelligence: DECIDE-
AI. Nature Medicine, 28(5), 924–933. https://doi.org/10.1038/s41591-022-01772-9
27. Vollmer, S., Mateen, B. A., Bohner, G., Király, F. J., Ghani, R., Jonsson, P., Cumbers, S., Jonas, A.,
McAllister, K. S. L., Myles, P., Granger, D., Birse, M., Branson, R., Moons, K. G. M., Collins, G. S.,
Ioannidis, J. P. A., Holmes, C., & Hemingway, H. (2020). Machine learning and artificial intelligence
research for patient benefit: 20 critical questions on transparency, replicability, ethics, and
effectiveness. The BMJ, 368, l6927. https://doi.org/10.1136/bmj.l6927
28. Willemink, M. J., Koszek, W. A., Hardell, C., Wu, J., Fleischmann, D., Harvey, H., Folio, L. R.,
Summers, R. M., Rubin, D. L., & Lungren, M. P. (2020). Preparing medical imaging data for machine
learning. Radiology, 295(1), 4–15. https://doi.org/10.1148/radiol.2020192224
Page 559
www.rsisinternational.org
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XV, Issue VI, June 2026
29. World Health Organization. (2021). Ethics and governance of artificial intelligence for health: WHO
guidance. https://www.who.int/publications/i/item/9789240029200
