An ontology-based approach for developing a harmonised data-validation tool for European cancer registration

Ontology-Based AI Design Patterns and Constraints in Cancer Registry Data Validation

Data validation in cancer registration is a critical operation but is resource-intensive and has traditionally depended on proprietary software. Ontology-based AI is a novel approach utilising machine reasoning based on axioms formally described in description logic. This is a different approach from deep learning AI techniques but not exclusive of them. The advantage of the ontology approach lies in its ability to address a number of challenges concurrently. The disadvantages relate to computational costs, which increase with language expressivity and the size of data sets, and class containment restrictions imposed by description logics. Both these aspects would benefit from the availability of design patterns, which is the motivation behind this study. We modelled the European cancer registry data validation rules in description logic using a number of design patterns and showed the viability of the approach. Reasoning speeds are a limiting factor for large cancer registry data sets comprising many hundreds of thousands of records, but these can be offset to a certain extent by developing the ontology in a modular way. Data validation is also a highly parallelisable process. Important potential future work in this domain would be to identify and optimise reusable design patterns, paying particular attention to avoiding any unintended reasoning efficiency hotspots.

An ontology design for validating childhood cancer registry data

Ontologies can provide a valuable role in the work of cancer registration, particularly as a tool for managing and navigating the various classification systems and coding rules. Further advantages accrue from the ability to formalise the coding rule base using description logics and thereby benefit from the associated automatic reasoning functionality. Drawing from earlier work that showed the viability of applying ontologies in the data validation tasks of cancer registries, an ontology was created using a modular approach to handle the specific checks for childhood cancers. The ontology was able to handle successfully the various inter-variable checks using the axiomatic constructs of the web ontology language. Application of an ontological approach for data validation can greatly simplify the maintenance of the coding rules and facilitate the federation of any centralised validation process to the local level. It also provides an improved means of visualising the rule interdependencies from different perspectives. Performance of the automatic reasoning process can be a limiting issue for very large datasets and will be a focus for future work. Results are provided showing how the ontology is able to validate cancer case records typical for childhood tumours.

Data Quality Check in Cancer Imaging Research: Deploying and Evaluating the DIQCT Tool

Data harmonization is one of the greatest challenges in cancer imaging studies, especially when it comes to multi-source data provision. Properly integrated data deriving from various sources can ensure data fairness on one side and can lead to a trusted dataset that will enhance AI engine development on the other side. Towards this direction, we are presenting a data integration quality check tool that ensures that all data uploaded to the repository are homogenized and share the same principles. The tool's aim is to report any human-induced errors and propose corrective actions. It focuses on checking the data prior to their upload to the repository in five levels: (i) clinical metadata integrity, (ii) template-imaging consistency, (iii) anonymization protocol applied, (iv) imaging analysis requirements, (v) case completeness. The tool produces reports with the corrective actions that must be followed by the user. This way the tool ensures that the data that will become available to the developers of the AI engine are homogenized, properly structured and contain all the necessary information needed for the analysis. The tool was validated in two rounds, internal and external, and at the user experience level. Clinical Relevance- Supporting the harmonized preparation and provision of medical imaging data and related clinical data will ensure data fairness and enhance the AI development.

Ontologies and Knowledge Graphs in Oncology Research

The complexity of cancer research stems from leaning on several biomedical disciplines for relevant sources of data, many of which are complex in their own right. A holistic view of cancer—which is critical for precision medicine approaches—hinges on integrating a variety of heterogeneous data sources under a cohesive knowledge model, a role which biomedical ontologies can fill. This study reviews the application of ontologies and knowledge graphs in cancer research. In total, our review encompasses 141 published works, which we categorized under 14 hierarchical categories according to their usage of ontologies and knowledge graphs. We also review the most commonly used ontologies and newly developed ones. Our review highlights the growing traction of ontologies in biomedical research in general, and cancer research in particular. Ontologies enable data accessibility, interoperability and integration, support data analysis, facilitate data interpretation and data mining, and more recently, with the emergence of the knowledge graph paradigm, support the application of Artificial Intelligence methods to unlock new knowledge from a holistic view of the available large volumes of heterogeneous data.

A multipurpose TNM stage ontology for cancer registries

Background

Population-based cancer registries are a critical reference source for the surveillance and control of cancer. Cancer registries work extensively with the internationally recognised TNM classification system used to stage solid tumours, but the system is complex and compounded by the different TNM editions in concurrent use. TNM ontologies exist but the design requirements are different for the needs of the clinical and cancer-registry domains. Two TNM ontologies developed specifically for cancer registries were designed for different purposes and have limitations for serving wider application. A unified ontology is proposed to serve the various cancer registry TNM-related tasks and reduce the multiplication effects of different ontologies serving specific tasks. The ontology is comprehensive of the rules for TNM edition 7 as required by cancer registries and designed on a modular basis to allow extension to other TNM editions.

Results

A unified ontology was developed building on the experience and design of the existing ontologies. It follows a modular approach allowing plug in of components dependent upon any particular TNM edition. A Java front-end was developed to interface with the ontology via the Web Ontology Language application programme interface and enables batch validation or classification of cancer registry records. The programme also allows the means of automated error correction in some instances. Initial tests verified the design concept by correctly inferring TNM stage and successfully handling the TNM-related validation checks on a number of cancer case records, with a performance similar to that of an existing ontology dedicated to the task.

Conclusions

The unified ontology provides a multi-purpose tool for TNM-related tasks in a cancer registry and is scalable for different editions of TNM. It offers a convenient way of quickly checking validity of cancer case stage information and for batch processing of multi-record data via a dedicated front-end programme. The ontology is adaptable to many uses, either as a standalone TNM module or as a component in applications of wider focus. It provides a first step towards a single, unified TNM ontology for cancer registries.