Generalization of deep learning models for ultra-low-count amyloid PET/MRI using transfer learning.
Purpose: We aimed to evaluate the performance of deep learning-based generalization of ultra-low-count amyloid PET/MRI enhancement when applied to studies acquired with different scanning hardware and protocols. Methods: Eighty simultaneous [18F]florbetaben PET/MRI studies were acquired, split equal...
| Publicado en: | European Journal of Nuclear Medicine & Molecular Imaging Vol. 47; no. 13; pp. 2998 - 3008 |
|---|---|
| Autores principales: | , , , , , , , , , , |
| Formato: | Journal Article |
| Publicado: |
Springer Nature
2020
|
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=147137016&site=ehost-live header: @attributes: shortDbName: ccm uiTerm: 147137016 longDbName: CINAHL Complete uiTag: AN controlInfo: bkinfo: dissinfo: jinfo: jid: 16197070 NPC jtl: European Journal of Nuclear Medicine & Molecular Imaging issn: 16197070 maglogo: N pubinfo: dt: 2020 vid: 47 iid: 13 pid: 237 pub: Springer Nature place: New York, New York artinfo: ui: 147137016 144035128 10.1007/s00259-020-04897-6 147137016 ppf: 2998 ppct: 10 formats: fmt: – @attributes: type: T – @attributes: type: P tig: atl: Generalization of deep learning models for ultra-low-count amyloid PET/MRI using transfer learning. aug: au: Chen, Kevin T. Schürer, Matti Ouyang, Jiahong Koran, Mary Ellen I. Davidzon, Guido Mormino, Elizabeth Tiepolt, Solveig Hoffmann, Karl-Titus Sabri, Osama Zaharchuk, Greg Barthel, Henryk affil: Department of Radiology, Stanford University, Stanford, CA, United States sug: ab: Purpose: We aimed to evaluate the performance of deep learning-based generalization of ultra-low-count amyloid PET/MRI enhancement when applied to studies acquired with different scanning hardware and protocols. Methods: Eighty simultaneous [18F]florbetaben PET/MRI studies were acquired, split equally between two sites (site 1: Signa PET/MRI, GE Healthcare, 39 participants, 67 ± 8 years, 23 females; site 2: mMR, Siemens Healthineers, 64 ± 11 years, 23 females) with different MRI protocols. Twenty minutes of list-mode PET data (90–110 min post-injection) were reconstructed as ground-truth. Ultra-low-count data obtained from undersampling by a factor of 100 (site 1) or the first minute of PET acquisition (site 2) were reconstructed for ultra-low-dose/ultra-short-time (1% dose and 5% time, respectively) PET images. A deep convolution neural network was pre-trained with site 1 data and either (A) directly applied or (B) trained further on site 2 data using transfer learning. Networks were also trained from scratch based on (C) site 2 data or (D) all data. Certified physicians determined amyloid uptake (+/−) status for accuracy and scored the image quality. The peak signal-to-noise ratio, structural similarity, and root-mean-squared error were calculated between images and their ground-truth counterparts. Mean regional standardized uptake value ratios (SUVR, reference region: cerebellar cortex) from 37 successful site 2 FreeSurfer segmentations were analyzed. Results: All network-synthesized images had reduced noise than their ultra-low-count reconstructions. Quantitatively, image metrics improved the most using method B, where SUVRs had the least variability from the ground-truth and the highest effect size to differentiate between positive and negative images. Method A images had lower accuracy and image quality than other methods; images synthesized from methods B–D scored similarly or better than the ground-truth images. Conclusions: Deep learning can successfully produce diagnostic amyloid PET images from short frame reconstructions. Data bias should be considered when applying pre-trained deep ultra-low-count amyloid PET/MRI networks for generalization. pubtype: Academic Journal doctype: Journal Article ougenre: Article language: English refInfo: holdings: @attributes: islocal: N |
|---|