Ontology-Driven Knowledge Sharing in Alzheimer’s Disease Research

Round 1

Reviewer 1 Report

Data and knowledge sharing is a highly relevant and important challenge, particularly in the health domain.

The work contributes in an interesting direction of ontology adoption in the health information systems, particularly for knowledge sharing in relation to Alzheimer disease.

The paper however needs improvements to be published.

There is little clarity at the start who is the audience of the designed ontology, or what the exact research question is. Is there an expectation that regular users should be able to interact with ontology in Protégé, and is it this point that was mainly tested here? Research question, hypotheses, starting assumptions and motivating scenarios should be specified explicitly.

It should be explained wherefrom the competency questions come, and why they are typical or characteristic. Now they look somewhat randomly created. Why they are also sufficient, quantity-wise?

When describing the selected users that are to take part in the study, it is unclear how they were chosen what their profiles exactly were, and what they should have been achieving in relation to research question.

Their profiles are also not described in detail, and are even presented ambiguously. For example, there is a statement about users that “ 80% of them reported having an occupational background in science with half of them having an interdisciplinary background in science and one or two other STEM fields.” – here and in the table the segmentation is not clear – what is meant by “science”? Mathematics, technology and engineering are also sciences – these are not disjunctive concepts as presented in the table.

In results section, there are many details relevant to Alzheimer disease and diagnostics, but it is not clear how this information is related (or not) or represented (or not) in the ontology, and how it is relevant to the study. There should be a clearer link from the domain to the technical developments.

The developed ontology should be made available for reuse via some repository e.g. Github and linked to in the paper. Otherwise others cannot examine and re-use the ontology, and the reported study is not reproducible.

The developed ontology is the main product of this work. However, the related works in references stem almost exclusively from Alzheimer related health studies (apart from very basic references on ontologies in general). Up-to-date practices and developments in ontology and semantic technology use for data and knowledge sharing should be reflected on to a due extent, also with the referenced literature, especially focusing on such data and knowledge sharing for medical/health domains. The relation of the current work to the whole ontology-relevant landscape here does not seem to be complete – related works on the technical side are not outlined in full, and explanations of what exactly was used from them and why are not provided.

There is a mention in the paper (text and diagrams) that clinical data are also exchanged and shared. However, these data are often connected to specific patients (personal data), for which consent should be received. This should be also elaborated in the ontology or as a minimum discussed. There are also semantic solutions and ontologies covering legal aspects of data sharing and data governance (consent, GDPR). Some of relevant example works to refer here are:

Kurteva, A., Chhetri, T. R., Pandit, H. J., & Fensel, A. (2021). Consent through the lens of semantics: State of the art survey and best practices. Semantic Web, 1-27.

Chhetri, T. R., Kurteva, A., DeLong, R. J., Hilscher, R., Korte, K., & Fensel, A. (2022). Data Protection by Design Tool for Automated GDPR Compliance Verification Based on Semantically Modeled Informed Consent. Sensors, 22(7), 2763.

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Title:

Ontology-Driven Knowledge Sharing in Alzheimer's Disease Research

 

Summary:

The work describes the development process of an ontology for the diagnosis and preclinical classification of Alzheimer's disease. The main purpose of such an ontology is the effective knowledge sharing in multidisciplinary teams working on the Alzheimer's disease. The ontology cover six major conceptual groups, i.e. the pathology, spectrum, diagnostic process, symptoms, assessments, and relevant clinical findings of the disease. The quality of the developed ontology is finally evaluated in terms of its usability, applicability, and correctness.

 

Comments:

The paper is well written and it follows a clear and reasonable structure. The methodology is nicely described and it is well rooted with the literature. I agree with the authors that one of the most important future work is the evaluation of this ontology by both a bigger sample, and by ontology experts. It is not clear to me whether the ontology developed has been released and it is usable also by other groups that might be interested in working with similar topics. Furthermore, at least for the purpose of reviewing the soundness of the work, I would have preferred to actually see the ontology, either as an appendix to the work, or as a .owl file attached to the submission.

 

The actual expressivity of the language used to describe the ontology has not been discussed, and I believe it is an important technical part. It is said that the ontology has been developed using OWL (I believe OWL 2.0). But what is the dialect used and/or needed to represent the ontology? (This information is available in Protégé). Also, is the ontology also meant to perform automatic reasoning on top of it, and if it is so, what kind of interactions one might expect from the usage of this ontology? Is the ontology reasonably fixed after this development phase, or do you expect it to be expanded in the future?

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

The paper presents an important effort to develop an ontology as a basic "common ground" tool facilitating the cooperation of different scientific disciplines experts to A.D related reserch.

It is very well written and presented providing all the necessary background information.

A weak point of the evalution is that it is based on ten persons only. However, it is mentioned in the ending that a future aim is to work with bigger and more diverse samples. As long as the initial evaluation and presentation of the system is regarded this is quite fine.

As the initials CQ based Q1 to Q6 are used and Q1/Q2 referring to scenario based also exist, a possible clarifying move is to name the right panel Q1 and Q2 as SQ1 and SQ2  (Scenario Questionarries).

Author Response

Please see the attachment.

Author Response File: Author Response.docx