search

aapm-bdsc-ontology-tg / radont

Apache License 2.0
3 stars 0 forks source link

Inference example #70

Open Bayesianworld opened 2 years ago

Bayesianworld commented 2 years ago

Now that we have a consistent ontology in owl, i.e. Reasoner doesn't throw errors, I want to devise an example that exhibits utility of inference. I have defined 4 subclasses of "technique"

SRS: single fraction (?), in brain SBRT: <= 5 Fx, uses MLC, dose > 6 Gy, inverse planning IMRT: >15 Fx or dose < 6 Gy, uses MLC, inverse planning hypofractionation: 6<Fx<16, can use forward or inverse planning

Will need to add a few things, e.g. brain. See next issue

andredekker commented 2 years ago

One of the things to consider is whether or not this should be in the RTO (as a domain ontology) or in an application ontology. E.g. what is and is not hypo-fractionation may not be consistent across sites/countries and also may change over time.

jonathanbona commented 2 years ago

I'd vote to flesh out these use cases in an "application ontology" file that imports the main RTO OWL file. This approach will also allow us to avoid pulling anatomy classes into RTO proper.

On Wed, Jun 29, 2022, 04:13 Andre Dekker @.***> wrote:

One of the things to consider is whether or not this should be in the RTO (as a domain ontology) or in an application ontology. E.g. what is and is not hypo-fractionation may not be consistent across sites/countries and also may change over time.

— Reply to this email directly, view it on GitHub https://github.com/aapm-bdsc-ontology-tg/radont/issues/70#issuecomment-1169729874, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAATBNZDRCTMEYCAPGXSCQDVRQHS5ANCNFSM52D4QBIQ . You are receiving this because you are subscribed to this thread.Message ID: @.***>

Bayesianworld commented 2 years ago

I agree totally, I was just interested in doing an interesting case for illustration purposes for the paper. I would like to also illustrate the merging of an application ontology and RTO, but I don't think OORO is written in OWL. Is ROO appropriate for such a test? Mark

From: Jonathan Bona @.> Sent: Wednesday, June 29, 2022 4:35 AM To: aapm-bdsc-ontology-tg/radont @.> Cc: Mark H. Phillips @.>; Author @.> Subject: Re: [aapm-bdsc-ontology-tg/radont] Inference example (Issue #70)

I'd vote to flesh out these use cases in an "application ontology" file that imports the main RTO OWL file. This approach will also allow us to avoid pulling anatomy classes into RTO proper.

On Wed, Jun 29, 2022, 04:13 Andre Dekker @.***> wrote:

One of the things to consider is whether or not this should be in the RTO (as a domain ontology) or in an application ontology. E.g. what is and is not hypo-fractionation may not be consistent across sites/countries and also may change over time.

— Reply to this email directly, view it on GitHub https://github.com/aapm-bdsc-ontology-tg/radont/issues/70#issuecomment-1169729874, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAATBNZDRCTMEYCAPGXSCQDVRQHS5ANCNFSM52D4QBIQ . You are receiving this because you are subscribed to this thread.Message ID: @.***>

— Reply to this email directly, view it on GitHubhttps://github.com/aapm-bdsc-ontology-tg/radont/issues/70#issuecomment-1169872826, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AGU5EEWT2Y7WYCTWS67F2BLVRQYJXANCNFSM52D4QBIQ. You are receiving this because you authored the thread.Message ID: @.***>