medizininformatik-initiative / kerndatensatzmodul-medikation

Kerndatensatzmodul Medikation
1 stars 0 forks source link

CodeSystem "UNII"/"FDASIS" is missing #62

Open jpwiedekopf opened 9 months ago

jpwiedekopf commented 9 months ago

The CodeSystem "UNII"/"FDASIS" (canonical http://fdasis.nlm.nih.gov) is not represented in the package and not available in FHIR format on the Internet.

A strategy for converting the CodeSystem from the native format, or a placeholder CodeSystem should be created should converting the CodeSystem not be possible due to technical or license reasons (in this case, the CodeSystem and the ImplementationGuide should describe how to acquire the master file to implementers).

jpwiedekopf commented 9 months ago

Assuming I did not misidentify http://fdasis.nlm.nih.gov as UNII, the master file is available free-of-charge and without copyright from the FDA: https://precision.fda.gov/uniisearch/archive

Creating a FHIR CS should be easy.

julsas commented 8 months ago

https://precision.fda.gov/uniisearch/archive

julsas commented 8 months ago

INN: https://www.wcoomd.org/en/topics/nomenclature/instrument-and-tools/tools-to-assist-with-the-classification-in-the-hs/hs_classification-decisions/inn-table.aspx

julsas commented 8 months ago

PZN: https://www2.bfarm.de/PZN/PZN.txt

julsas commented 8 months ago

ASK und CAS: https://www.dimdi.de/dynamic/de/arzneimittel/downloads/

DannyAmmon commented 8 months ago

PZN: https://www2.bfarm.de/PZN/PZN.txt

Das sind aber nur BTM?

jpwiedekopf commented 8 months ago

PZN: https://www.wido.de/publikationen-produkte/analytik/arzneimittel-klassifikation/arzneimittel-stammdatei/?L=0 (???)

julsas commented 8 months ago

Das sind aber nur BTM?

Mist, sieht so aus.

jpwiedekopf commented 8 months ago

ASK und CAS: https://www.dimdi.de/dynamic/de/arzneimittel/downloads/

Cool, wusste nicht dass die public sind. Eine große CSV-Datei mit sowohl ASK als auch CAS ergibt sicher Sinn, aber daraus sollte vermutlich nur ASK generiert werden, oder CAS maximal mit content=fragment, weil CAS Codes für alle möglichen chemischen Verbindungen generiert, nicht nur für pharmazeutische Wirkstoffe (AFAIK). Konvertieren könnte recht gut mit BabelFSH gehen, aber es fehlt momentan noch die Möglichkeit, Designations zu generieren, das könnte hier relevant sein. Alternativ sind viele Spalten auch als Properties abzubilden.

Wenn jemand von euch Lust hat, einen Aufschlag für eine FSH-Datei zu schreiben, gerne, ich schaffe es erst wieder in der ersten vollen März-Woche.

UNII (https://precision.fda.gov/uniisearch/archive)

Sollte sogar noch einfacher mit BabelFSH sein.

beakmachine commented 8 months ago

Also, wir können sinnvolle Mappings ASK ⇔ UNII ⇔ CAS zur Verfügung stellen.

Wir brauchen eigentlich wirklich keine Liste von PZN. Das wird nicht zur Suche verwendet.

Allerdings sollte das BfArM schon vor zwei Jahren eine derartige Liste/db mit entsprechenden IDMP-konformen Informationen bereit gestellt haben! Hat es aber nicht, wie es andere Sachen auch nicht haben wird.

jpwiedekopf commented 3 months ago

ASK und CAS: https://www.dimdi.de/dynamic/de/arzneimittel/downloads/

Cool, wusste nicht dass die public sind. Eine große CSV-Datei mit sowohl ASK als auch CAS ergibt sicher Sinn, aber daraus sollte vermutlich nur ASK generiert werden, oder CAS maximal mit content=fragment, weil CAS Codes für alle möglichen chemischen Verbindungen generiert, nicht nur für pharmazeutische Wirkstoffe (AFAIK). Konvertieren könnte recht gut mit BabelFSH gehen, aber es fehlt momentan noch die Möglichkeit, Designations zu generieren, das könnte hier relevant sein. Alternativ sind viele Spalten auch als Properties abzubilden.

Mittlerweile unter https://www.bfarm.de/DE/Arzneimittel/Arzneimittelinformationen/Arzneimittel-recherchieren/Stoffbezeichnungen/_node.html

julsas commented 1 week ago

PZN: https://www.wido.de/publikationen-produkte/analytik/arzneimittel-klassifikation/arzneimittel-stammdatei/?L=0 (???)

Habe da mal angefragt, wie es mit der Lizensierung der Stammdatei für die MII aussieht.