From histopathology images to molecular characterisation of
tumours: The artificial intelligence path

C 2025

From histopathology images to molecular characterisation of tumours: The artificial intelligence path

POPOVICI, Vlad a Daniel RACOCEANU

Základní údaje

Originální název

From histopathology images to molecular characterisation of tumours: The artificial intelligence path

Autoři

POPOVICI, Vlad a Daniel RACOCEANU

Vydání

New Delhi, Recent Advances in Histopathology, od s. 1-21, 21 s. vol. 27, 2025

Nakladatel

Jaypee Brothers Medical Publishers

Další údaje

Jazyk

angličtina

Typ výsledku

Kapitola resp. kapitoly v odborné knize

Obor

30109 Pathology

Stát vydavatele

Indie

Utajení

není předmětem státního či obchodního tajemství

Forma vydání

tištěná verze "print"

Odkazy

URL

Označené pro přenos do RIV

Ano

Kód RIV

RIV/00216224:14310/25:00143732

Organizační jednotka

Přírodovědecká fakulta

ISBN

978-1-78779-183-1

Klíčová slova anglicky

computational pathology; histopathology; artificial intelligence

Štítky

143180, rivok

Příznaky

Mezinárodní význam

Změněno: 2. 4. 2026 14:08, Mgr. Marie Novosadová Šípková, DiS.

Anotace

V originále

Artificial neural networks regained the spotlight of AI research efforts and are now the main driving force behind spectacular developments in a wide range of domains. For medical imaging in general and pathological imaging in particular, these advancements open a whole new universe of applications. Molecular data represents a natural companion to imaging and the results briefly reviewed in this chapter demonstrate some of the possibilities of such explorations. Currently, apart from MSI endpoint, the performance of the models predicting molecular features, while promising, does not warrant their application in clinical scenarios. This does not only necessarily indicate a limitation of the network architectures employed, but also issues with the data used in training the models. Transitioning the AI models for pathology from research laboratory to clinical practice still requires extensive validation and faces several major challenges from more technical (performance and stability across platforms) to interpretability and regulatory. But, the question is not whether these AI companions will become ubiquitous, but rather ‘when?’.

Návaznosti

LM2023069, projekt VaV

Název: Výzkumná infrastruktura RECETOX

Investor: Ministerstvo školství, mládeže a tělovýchovy ČR, Výzkumná infrastruktura RECETOX

Citovat

POPOVICI, Vlad a Daniel RACOCEANU. From histopathology images to molecular characterisation of tumours: The artificial intelligence path. In Fred T Bosman; Ivan Damjanov. Recent Advances in Histopathology. New Delhi: Jaypee Brothers Medical Publishers, 2025, s. 1-21. vol. 27. ISBN 978-1-78779-183-1.

@inbook{2555086,
   author = {Popovici, Vlad and Racoceanu, Daniel},
   address = {New Delhi},
   booktitle = {Recent Advances in Histopathology},
   editor = {Fred T Bosman; Ivan Damjanov},
   keywords = {computational pathology; histopathology; artificial intelligence},
   howpublished = {tištěná verze "print"},
   language = {eng},
   location = {New Delhi},
   isbn = {978-1-78779-183-1},
   pages = {1-21},
   publisher = {Jaypee Brothers Medical Publishers},
   title = {From histopathology images to molecular characterisation of tumours: The artificial intelligence path},
   url = {https://hal.science/hal-04895755v1/file/popovici-racoceanu2025histopathology_to_molecular.pdf},
   year = {2025}
}

TY  - CHAP
ID  - 2555086
AU  - Popovici, Vlad - Racoceanu, Daniel
PY  - 2025
TI  - From histopathology images to molecular characterisation of tumours: The artificial intelligence path
VL  - vol. 27
PB  - Jaypee Brothers Medical Publishers
CY  - New Delhi
SN  - 9781787791831
KW  - computational pathology
KW  - histopathology
KW  - artificial intelligence
UR  - https://hal.science/hal-04895755v1/file/popovici-racoceanu2025histopathology_to_molecular.pdf
N2  - Artificial neural networks regained the spotlight of AI research efforts and are now the main driving force behind spectacular developments in a wide range of domains. For medical imaging in general and pathological imaging in particular, these advancements open a whole new universe of applications. Molecular data represents a natural companion to imaging and the results briefly reviewed in this chapter demonstrate some of the possibilities of such explorations. Currently, apart from MSI endpoint, the performance of the models predicting molecular features, while promising, does not warrant their application in clinical scenarios. This does not only necessarily indicate a limitation of the network architectures employed, but also issues with the data used in training the models. Transitioning the AI models for pathology from research laboratory to clinical practice still requires extensive validation and faces several major challenges from more technical (performance and stability across platforms) to interpretability and regulatory. But, the question is not whether these AI companions will become ubiquitous, but rather ‘when?’.
ER  -

POPOVICI, Vlad a Daniel RACOCEANU. From histopathology images to molecular characterisation of tumours: The artificial intelligence path. In Fred T Bosman; Ivan Damjanov. \textit{Recent Advances in Histopathology}. New Delhi: Jaypee Brothers Medical Publishers, 2025, s.~1-21. vol. 27. ISBN~978-1-78779-183-1.

Přehled o publikaci