Automatic etiological classification of stroke thrombus digital photographs using a deep learning model
BackgroundEtiological classification of ischemic stroke is fundamental for secondary prevention, but frequently results in undetermined cause. We aimed to develop a Deep Learning (DL)-based model for automatic etiological classification of ischemic stroke using digital images of thrombi retrieved by...
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-01-01
|
| Series: | Frontiers in Neurology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fneur.2025.1534845/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841533216250396672 |
|---|---|
| author | Álvaro Lucero-Garófano Álvaro Lucero-Garófano Alicia Aliena-Valero Isabel Vielba-Gómez Isabel Vielba-Gómez Irene Escudero-Martínez Irene Escudero-Martínez Lluís Morales-Caba Lluís Morales-Caba Fernando Aparici-Robles Fernando Aparici-Robles Diana L. Tarruella Hernández Diana L. Tarruella Hernández Gerardo Fortea Gerardo Fortea José I. Tembl José I. Tembl Juan B. Salom Juan B. Salom José V. Manjón |
| author_facet | Álvaro Lucero-Garófano Álvaro Lucero-Garófano Alicia Aliena-Valero Isabel Vielba-Gómez Isabel Vielba-Gómez Irene Escudero-Martínez Irene Escudero-Martínez Lluís Morales-Caba Lluís Morales-Caba Fernando Aparici-Robles Fernando Aparici-Robles Diana L. Tarruella Hernández Diana L. Tarruella Hernández Gerardo Fortea Gerardo Fortea José I. Tembl José I. Tembl Juan B. Salom Juan B. Salom José V. Manjón |
| author_sort | Álvaro Lucero-Garófano |
| collection | DOAJ |
| description | BackgroundEtiological classification of ischemic stroke is fundamental for secondary prevention, but frequently results in undetermined cause. We aimed to develop a Deep Learning (DL)-based model for automatic etiological classification of ischemic stroke using digital images of thrombi retrieved by mechanical thrombectomy.MethodsPatients with large vessel occlusion stroke subjected to mechanical thrombectomy between April 2016 and January 2023 at La Fe University and Polytechnic Hospital in Valencia were included. Thrombus digital images were obtained and clinical characteristics, including TOAST etiological classification as reference standard, were retrieved. Statistical analysis was performed to compare clinical characteristics between atherothrombotic and cardioembolic strokes. A DL method was designed based on two deep neural networks for: (1) image segmentation and (2) image classification including clinical characteristics. The metrics used were DICE coefficient for the segmentation network, and accuracy, precision, sensitivity, specificity and area under the curve (AUC) for the predictions of the classification network.ResultsA total of 166 patients (mean age 69 [SD, 13], 67 female) were included. TOAST classification was: 31 atherothrombotic, 87 cardioembolic, and 48 cryptogenic. The segmentation network achieved an average DICE coefficient of 0.96 [SD, 0.13]. The optimal fused imaging and clinical classification network had a 0.968 accuracy [95% CI, 0.935–0.994], and AUC of 0.947 [95% CI, 0.870–1]. Cryptogenic thrombi were classified as cardioembolic (96%) or atherothrombotic (4%).ConclusionTwo convolutional neural networks perform the automatic segmentation of thrombus images and, combined with selected clinical characteristics, their accurate and precise classification into atherothrombotic or cardioembolic etiology in patients with acute ischemic stroke. |
| format | Article |
| id | doaj-art-1f49365e44dc4770b0fb500df3b267b8 |
| institution | Kabale University |
| issn | 1664-2295 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neurology |
| spelling | doaj-art-1f49365e44dc4770b0fb500df3b267b82025-01-17T05:10:43ZengFrontiers Media S.A.Frontiers in Neurology1664-22952025-01-011610.3389/fneur.2025.15348451534845Automatic etiological classification of stroke thrombus digital photographs using a deep learning modelÁlvaro Lucero-Garófano0Álvaro Lucero-Garófano1Alicia Aliena-Valero2Isabel Vielba-Gómez3Isabel Vielba-Gómez4Irene Escudero-Martínez5Irene Escudero-Martínez6Lluís Morales-Caba7Lluís Morales-Caba8Fernando Aparici-Robles9Fernando Aparici-Robles10Diana L. Tarruella Hernández11Diana L. Tarruella Hernández12Gerardo Fortea13Gerardo Fortea14José I. Tembl15José I. Tembl16Juan B. Salom17Juan B. Salom18José V. Manjón19Unidad Mixta de Investigación Cerebrovascular, Instituto de Investigación Sanitaria La Fe, Valencia, SpainInstituto de Aplicaciones de las Tecnologías de la Información y de las Comunicaciones Avanzadas (ITACA), Universitat Politècnica de València, Valencia, SpainUnidad Mixta de Investigación Cerebrovascular, Instituto de Investigación Sanitaria La Fe, Valencia, SpainUnidad Mixta de Investigación Cerebrovascular, Instituto de Investigación Sanitaria La Fe, Valencia, SpainUnidad de Ictus, Servicio de Neurología, Hospital Universitario y Politécnico La Fe, Valencia, SpainUnidad Mixta de Investigación Cerebrovascular, Instituto de Investigación Sanitaria La Fe, Valencia, SpainUnidad de Ictus, Servicio de Neurología, Hospital Universitario y Politécnico La Fe, Valencia, SpainUnidad Mixta de Investigación Cerebrovascular, Instituto de Investigación Sanitaria La Fe, Valencia, SpainUnidad de Ictus, Servicio de Neurología, Hospital Universitario y Politécnico La Fe, Valencia, SpainUnidad Mixta de Investigación Cerebrovascular, Instituto de Investigación Sanitaria La Fe, Valencia, SpainServicio de Radiología, Hospital Universitario y Politécnico La Fe, Valencia, SpainUnidad Mixta de Investigación Cerebrovascular, Instituto de Investigación Sanitaria La Fe, Valencia, SpainUnidad de Ictus, Servicio de Neurología, Hospital Universitario y Politécnico La Fe, Valencia, SpainUnidad Mixta de Investigación Cerebrovascular, Instituto de Investigación Sanitaria La Fe, Valencia, SpainUnidad de Ictus, Servicio de Neurología, Hospital Universitario y Politécnico La Fe, Valencia, SpainUnidad Mixta de Investigación Cerebrovascular, Instituto de Investigación Sanitaria La Fe, Valencia, SpainUnidad de Ictus, Servicio de Neurología, Hospital Universitario y Politécnico La Fe, Valencia, SpainUnidad Mixta de Investigación Cerebrovascular, Instituto de Investigación Sanitaria La Fe, Valencia, SpainDepartamento de Fisiología, Universitat de València, Valencia, SpainInstituto de Aplicaciones de las Tecnologías de la Información y de las Comunicaciones Avanzadas (ITACA), Universitat Politècnica de València, Valencia, SpainBackgroundEtiological classification of ischemic stroke is fundamental for secondary prevention, but frequently results in undetermined cause. We aimed to develop a Deep Learning (DL)-based model for automatic etiological classification of ischemic stroke using digital images of thrombi retrieved by mechanical thrombectomy.MethodsPatients with large vessel occlusion stroke subjected to mechanical thrombectomy between April 2016 and January 2023 at La Fe University and Polytechnic Hospital in Valencia were included. Thrombus digital images were obtained and clinical characteristics, including TOAST etiological classification as reference standard, were retrieved. Statistical analysis was performed to compare clinical characteristics between atherothrombotic and cardioembolic strokes. A DL method was designed based on two deep neural networks for: (1) image segmentation and (2) image classification including clinical characteristics. The metrics used were DICE coefficient for the segmentation network, and accuracy, precision, sensitivity, specificity and area under the curve (AUC) for the predictions of the classification network.ResultsA total of 166 patients (mean age 69 [SD, 13], 67 female) were included. TOAST classification was: 31 atherothrombotic, 87 cardioembolic, and 48 cryptogenic. The segmentation network achieved an average DICE coefficient of 0.96 [SD, 0.13]. The optimal fused imaging and clinical classification network had a 0.968 accuracy [95% CI, 0.935–0.994], and AUC of 0.947 [95% CI, 0.870–1]. Cryptogenic thrombi were classified as cardioembolic (96%) or atherothrombotic (4%).ConclusionTwo convolutional neural networks perform the automatic segmentation of thrombus images and, combined with selected clinical characteristics, their accurate and precise classification into atherothrombotic or cardioembolic etiology in patients with acute ischemic stroke.https://www.frontiersin.org/articles/10.3389/fneur.2025.1534845/fullischemic strokeetiologyartificial intelligencedeep learningsegmentationclassification |
| spellingShingle | Álvaro Lucero-Garófano Álvaro Lucero-Garófano Alicia Aliena-Valero Isabel Vielba-Gómez Isabel Vielba-Gómez Irene Escudero-Martínez Irene Escudero-Martínez Lluís Morales-Caba Lluís Morales-Caba Fernando Aparici-Robles Fernando Aparici-Robles Diana L. Tarruella Hernández Diana L. Tarruella Hernández Gerardo Fortea Gerardo Fortea José I. Tembl José I. Tembl Juan B. Salom Juan B. Salom José V. Manjón Automatic etiological classification of stroke thrombus digital photographs using a deep learning model Frontiers in Neurology ischemic stroke etiology artificial intelligence deep learning segmentation classification |
| title | Automatic etiological classification of stroke thrombus digital photographs using a deep learning model |
| title_full | Automatic etiological classification of stroke thrombus digital photographs using a deep learning model |
| title_fullStr | Automatic etiological classification of stroke thrombus digital photographs using a deep learning model |
| title_full_unstemmed | Automatic etiological classification of stroke thrombus digital photographs using a deep learning model |
| title_short | Automatic etiological classification of stroke thrombus digital photographs using a deep learning model |
| title_sort | automatic etiological classification of stroke thrombus digital photographs using a deep learning model |
| topic | ischemic stroke etiology artificial intelligence deep learning segmentation classification |
| url | https://www.frontiersin.org/articles/10.3389/fneur.2025.1534845/full |
| work_keys_str_mv | AT alvarolucerogarofano automaticetiologicalclassificationofstrokethrombusdigitalphotographsusingadeeplearningmodel AT alvarolucerogarofano automaticetiologicalclassificationofstrokethrombusdigitalphotographsusingadeeplearningmodel AT aliciaalienavalero automaticetiologicalclassificationofstrokethrombusdigitalphotographsusingadeeplearningmodel AT isabelvielbagomez automaticetiologicalclassificationofstrokethrombusdigitalphotographsusingadeeplearningmodel AT isabelvielbagomez automaticetiologicalclassificationofstrokethrombusdigitalphotographsusingadeeplearningmodel AT ireneescuderomartinez automaticetiologicalclassificationofstrokethrombusdigitalphotographsusingadeeplearningmodel AT ireneescuderomartinez automaticetiologicalclassificationofstrokethrombusdigitalphotographsusingadeeplearningmodel AT lluismoralescaba automaticetiologicalclassificationofstrokethrombusdigitalphotographsusingadeeplearningmodel AT lluismoralescaba automaticetiologicalclassificationofstrokethrombusdigitalphotographsusingadeeplearningmodel AT fernandoaparicirobles automaticetiologicalclassificationofstrokethrombusdigitalphotographsusingadeeplearningmodel AT fernandoaparicirobles automaticetiologicalclassificationofstrokethrombusdigitalphotographsusingadeeplearningmodel AT dianaltarruellahernandez automaticetiologicalclassificationofstrokethrombusdigitalphotographsusingadeeplearningmodel AT dianaltarruellahernandez automaticetiologicalclassificationofstrokethrombusdigitalphotographsusingadeeplearningmodel AT gerardofortea automaticetiologicalclassificationofstrokethrombusdigitalphotographsusingadeeplearningmodel AT gerardofortea automaticetiologicalclassificationofstrokethrombusdigitalphotographsusingadeeplearningmodel AT joseitembl automaticetiologicalclassificationofstrokethrombusdigitalphotographsusingadeeplearningmodel AT joseitembl automaticetiologicalclassificationofstrokethrombusdigitalphotographsusingadeeplearningmodel AT juanbsalom automaticetiologicalclassificationofstrokethrombusdigitalphotographsusingadeeplearningmodel AT juanbsalom automaticetiologicalclassificationofstrokethrombusdigitalphotographsusingadeeplearningmodel AT josevmanjon automaticetiologicalclassificationofstrokethrombusdigitalphotographsusingadeeplearningmodel |