izpis_h1_title_alt

Extensive T1-weighted MRI preprocessing improves generalizability of deep brain age prediction models
ID Dular, Lara (Avtor), ID Pernuš, Franjo (Avtor), ID Špiclin, Žiga (Avtor)

.pdfPDF - Predstavitvena datoteka, prenos (4,29 MB)
MD5: DF3BF78F70967CD3E8D46FCAE10F042D
URLURL - Izvorni URL, za dostop obiščite https://www.sciencedirect.com/science/article/pii/S0010482524004049 Povezava se odpre v novem oknu

Izvleček
Brain age is an estimate of chronological age obtained from T1-weighted magnetic resonance images (T1w MRI), representing a straightforward diagnostic biomarker of brain aging and associated diseases. While the current best accuracy of brain age predictions on T1w MRIs of healthy subjects ranges from two to three years, comparing results across studies is challenging due to differences in the datasets, T1w preprocessing pipelines, and evaluation protocols used. This paper investigates the impact of T1w image preprocessing on the performance of four deep learning brain age models from recent literature. Four preprocessing pipelines, which differed in terms of registration transform, grayscale correction, and software implementation, were evaluated. The results showed that the choice of software or preprocessing steps could significantly affect the prediction error, with a maximum increase of 0.75 years in mean absolute error (MAE) for the same model and dataset. While grayscale correction had no significant impact on MAE, using affine rather than rigid registration to brain atlas statistically significantly improved MAE. Models trained on 3D images with isotropic 1 mm$^3$ resolution exhibited less sensitivity to the T1w preprocessing variations compared to 2D models or those trained on downsampled 3D images. Our findings indicate that extensive T1w preprocessing improves MAE, especially when predicting on a new dataset. This runs counter to prevailing research literature, which suggests that models trained on minimally preprocessed T1w scans are better suited for age predictions on MRIs from unseen scanners. We demonstrate that, irrespective of the model or T1w preprocessing used during training, applying some form of offset correction is essential to enable the model’s performance to generalize effectively on datasets from unseen sites, regardless of whether they have undergone the same or different T1w preprocessing as the training set.

Jezik:Angleški jezik
Ključne besede:magnetic resonance imaging, brain age prediction, image preprocessing, deep model regression, comparative study, quantitative evaluation, brain age, MRI preprocessing, deep regression models, linear mixed effect models, dataset bias, transfer learning, reproducible research, UK Biobank
Vrsta gradiva:Članek v reviji
Tipologija:1.01 - Izvirni znanstveni članek
Organizacija:FE - Fakulteta za elektrotehniko
Status publikacije:Objavljeno
Različica publikacije:Objavljena publikacija
Leto izida:2024
Št. strani:18 str.
Številčenje:Vol. 173, art. 108320
PID:20.500.12556/RUL-155885 Povezava se odpre v novem oknu
UDK:004.93:616.8
ISSN pri članku:1879-0534
DOI:10.1016/j.compbiomed.2024.108320 Povezava se odpre v novem oknu
COBISS.SI-ID:189802499 Povezava se odpre v novem oknu
Datum objave v RUL:23.04.2024
Število ogledov:199
Število prenosov:35
Metapodatki:XML RDF-CHPDL DC-XML DC-RDF
:
Kopiraj citat
Objavi na:Bookmark and Share

Gradivo je del revije

Naslov:Computers in biology and medicine
Skrajšan naslov:Comput. biol. & med.
Založnik:Elsevier
ISSN:1879-0534
COBISS.SI-ID:518726681 Povezava se odpre v novem oknu

Licence

Licenca:CC BY 4.0, Creative Commons Priznanje avtorstva 4.0 Mednarodna
Povezava:http://creativecommons.org/licenses/by/4.0/deed.sl
Opis:To je standardna licenca Creative Commons, ki daje uporabnikom največ možnosti za nadaljnjo uporabo dela, pri čemer morajo navesti avtorja.

Sekundarni jezik

Jezik:Slovenski jezik
Ključne besede:magnetnoresonančno slikanje, napovedovanje možganske starosti, preobdelava slik, globoki regresijski modeli, primerjalna študija, kvantitativno vrednotenje

Projekti

Financer:ARIS - Javna agencija za znanstvenoraziskovalno in inovacijsko dejavnost Republike Slovenije
Številka projekta:P2-0232
Naslov:Analiza biomedicinskih slik in signalov

Financer:ARIS - Javna agencija za znanstvenoraziskovalno in inovacijsko dejavnost Republike Slovenije
Številka projekta:J2-2500
Naslov:Analiza medicinskih slik s strojnim učenjem za napovedovanje poteka možganskih bolezni in učinkovitosti terapije

Financer:ARIS - Javna agencija za znanstvenoraziskovalno in inovacijsko dejavnost Republike Slovenije
Številka projekta:J2-3059
Naslov:Sprotno prilagajanje načrta protonske in radioterapije

Podobna dela

Podobna dela v RUL:
Podobna dela v drugih slovenskih zbirkah:

Nazaj