Convolutional neural networks for detection of transthyretin amyloidosis in 2D scintigraphy images
Abstract Background Transthyretin amyloidosis (ATTR) is a progressive disease which can be diagnosed non-invasively using bone avid [99mTc]-labeled radiotracers. Thus, ATTR is also an occasional incidental finding on bone scintigraphy. In this study, we trained convolutional neural networks (CNN) to...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2022-05-01
|
| Series: | EJNMMI Research |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s13550-022-00897-9 |
| _version_ | 1818191298249621504 |
|---|---|
| author | Hanna-Leena Halme Toni Ihalainen Olli Suomalainen Antti Loimaala Sorjo Mätzke Valtteri Uusitalo Outi Sipilä Eero Hippeläinen |
| author_facet | Hanna-Leena Halme Toni Ihalainen Olli Suomalainen Antti Loimaala Sorjo Mätzke Valtteri Uusitalo Outi Sipilä Eero Hippeläinen |
| author_sort | Hanna-Leena Halme |
| collection | DOAJ |
| description | Abstract Background Transthyretin amyloidosis (ATTR) is a progressive disease which can be diagnosed non-invasively using bone avid [99mTc]-labeled radiotracers. Thus, ATTR is also an occasional incidental finding on bone scintigraphy. In this study, we trained convolutional neural networks (CNN) to automatically detect and classify ATTR from scintigraphy images. The study population consisted of 1334 patients who underwent [99mTc]-labeled hydroxymethylene diphosphonate (HMDP) scintigraphy and were visually graded using Perugini grades (grades 0–3). A total of 47 patients had visual grade ≥ 2 which was considered positive for ATTR. Two custom-made CNN architectures were trained to discriminate between the four Perugini grades of cardiac uptake. The classification performance was compared to four state-of-the-art CNN models. Results Our CNN models performed better than, or equally well as, the state-of-the-art models in detection and classification of cardiac uptake. Both models achieved area under the curve (AUC) ≥ 0.85 in the four-class Perugini grade classification. Accuracy was good in detection of negative vs. positive ATTR patients (grade < 2 vs grade ≥ 2, AUC > 0.88) and high-grade cardiac uptake vs. other patients (grade < 3 vs. grade 3, AUC = 0.94). Maximum activation maps demonstrated that the automated deep learning models were focused on detecting the myocardium and not extracardiac features. Conclusion Automated convolutional neural networks can accurately detect and classify different grades of cardiac uptake on bone scintigraphy. The CNN models are focused on clinically relevant image features. Automated screening of bone scintigraphy images using CNN could improve the early diagnosis of ATTR. |
| first_indexed | 2024-12-12T00:12:23Z |
| format | Article |
| id | doaj.art-488ed193044b4d1285e1aa86d629aa65 |
| institution | Directory Open Access Journal |
| issn | 2191-219X |
| language | English |
| last_indexed | 2024-12-12T00:12:23Z |
| publishDate | 2022-05-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | EJNMMI Research |
| spelling | doaj.art-488ed193044b4d1285e1aa86d629aa652022-12-22T00:44:56ZengSpringerOpenEJNMMI Research2191-219X2022-05-0112111110.1186/s13550-022-00897-9Convolutional neural networks for detection of transthyretin amyloidosis in 2D scintigraphy imagesHanna-Leena Halme0Toni Ihalainen1Olli Suomalainen2Antti Loimaala3Sorjo Mätzke4Valtteri Uusitalo5Outi Sipilä6Eero Hippeläinen7Clinical Physiology and Nuclear Medicine, Helsinki University Hospital and University of HelsinkiClinical Physiology and Nuclear Medicine, Helsinki University Hospital and University of HelsinkiHeart and Lung Center, Helsinki University Hospital and University of HelsinkiClinical Physiology and Nuclear Medicine, Helsinki University Hospital and University of HelsinkiClinical Physiology and Nuclear Medicine, Helsinki University Hospital and University of HelsinkiClinical Physiology and Nuclear Medicine, Helsinki University Hospital and University of HelsinkiClinical Physiology and Nuclear Medicine, Helsinki University Hospital and University of HelsinkiClinical Physiology and Nuclear Medicine, Helsinki University Hospital and University of HelsinkiAbstract Background Transthyretin amyloidosis (ATTR) is a progressive disease which can be diagnosed non-invasively using bone avid [99mTc]-labeled radiotracers. Thus, ATTR is also an occasional incidental finding on bone scintigraphy. In this study, we trained convolutional neural networks (CNN) to automatically detect and classify ATTR from scintigraphy images. The study population consisted of 1334 patients who underwent [99mTc]-labeled hydroxymethylene diphosphonate (HMDP) scintigraphy and were visually graded using Perugini grades (grades 0–3). A total of 47 patients had visual grade ≥ 2 which was considered positive for ATTR. Two custom-made CNN architectures were trained to discriminate between the four Perugini grades of cardiac uptake. The classification performance was compared to four state-of-the-art CNN models. Results Our CNN models performed better than, or equally well as, the state-of-the-art models in detection and classification of cardiac uptake. Both models achieved area under the curve (AUC) ≥ 0.85 in the four-class Perugini grade classification. Accuracy was good in detection of negative vs. positive ATTR patients (grade < 2 vs grade ≥ 2, AUC > 0.88) and high-grade cardiac uptake vs. other patients (grade < 3 vs. grade 3, AUC = 0.94). Maximum activation maps demonstrated that the automated deep learning models were focused on detecting the myocardium and not extracardiac features. Conclusion Automated convolutional neural networks can accurately detect and classify different grades of cardiac uptake on bone scintigraphy. The CNN models are focused on clinically relevant image features. Automated screening of bone scintigraphy images using CNN could improve the early diagnosis of ATTR.https://doi.org/10.1186/s13550-022-00897-9AmyloidosisTransthyretinScintigraphyDeep learningConvolutional neural network |
| spellingShingle | Hanna-Leena Halme Toni Ihalainen Olli Suomalainen Antti Loimaala Sorjo Mätzke Valtteri Uusitalo Outi Sipilä Eero Hippeläinen Convolutional neural networks for detection of transthyretin amyloidosis in 2D scintigraphy images EJNMMI Research Amyloidosis Transthyretin Scintigraphy Deep learning Convolutional neural network |
| title | Convolutional neural networks for detection of transthyretin amyloidosis in 2D scintigraphy images |
| title_full | Convolutional neural networks for detection of transthyretin amyloidosis in 2D scintigraphy images |
| title_fullStr | Convolutional neural networks for detection of transthyretin amyloidosis in 2D scintigraphy images |
| title_full_unstemmed | Convolutional neural networks for detection of transthyretin amyloidosis in 2D scintigraphy images |
| title_short | Convolutional neural networks for detection of transthyretin amyloidosis in 2D scintigraphy images |
| title_sort | convolutional neural networks for detection of transthyretin amyloidosis in 2d scintigraphy images |
| topic | Amyloidosis Transthyretin Scintigraphy Deep learning Convolutional neural network |
| url | https://doi.org/10.1186/s13550-022-00897-9 |
| work_keys_str_mv | AT hannaleenahalme convolutionalneuralnetworksfordetectionoftransthyretinamyloidosisin2dscintigraphyimages AT toniihalainen convolutionalneuralnetworksfordetectionoftransthyretinamyloidosisin2dscintigraphyimages AT ollisuomalainen convolutionalneuralnetworksfordetectionoftransthyretinamyloidosisin2dscintigraphyimages AT anttiloimaala convolutionalneuralnetworksfordetectionoftransthyretinamyloidosisin2dscintigraphyimages AT sorjomatzke convolutionalneuralnetworksfordetectionoftransthyretinamyloidosisin2dscintigraphyimages AT valtteriuusitalo convolutionalneuralnetworksfordetectionoftransthyretinamyloidosisin2dscintigraphyimages AT outisipila convolutionalneuralnetworksfordetectionoftransthyretinamyloidosisin2dscintigraphyimages AT eerohippelainen convolutionalneuralnetworksfordetectionoftransthyretinamyloidosisin2dscintigraphyimages |