Screening for Chagas disease from the electrocardiogram using a deep neural network.
<h4>Background</h4>Worldwide, it is estimated that over 6 million people are infected with Chagas disease (ChD). It is a neglected disease that can lead to severe heart conditions in its chronic phase. While early treatment can avoid complications, the early-stage detection rate is low....
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2023-07-01
|
| Series: | PLoS Neglected Tropical Diseases |
| Online Access: | https://journals.plos.org/plosntds/article/file?id=10.1371/journal.pntd.0011118&type=printable |
| _version_ | 1797303880985346048 |
|---|---|
| author | Carl Jidling Daniel Gedon Thomas B Schön Claudia Di Lorenzo Oliveira Clareci Silva Cardoso Ariela Mota Ferreira Luana Giatti Sandhi Maria Barreto Ester C Sabino Antonio L P Ribeiro Antônio H Ribeiro |
| author_facet | Carl Jidling Daniel Gedon Thomas B Schön Claudia Di Lorenzo Oliveira Clareci Silva Cardoso Ariela Mota Ferreira Luana Giatti Sandhi Maria Barreto Ester C Sabino Antonio L P Ribeiro Antônio H Ribeiro |
| author_sort | Carl Jidling |
| collection | DOAJ |
| description | <h4>Background</h4>Worldwide, it is estimated that over 6 million people are infected with Chagas disease (ChD). It is a neglected disease that can lead to severe heart conditions in its chronic phase. While early treatment can avoid complications, the early-stage detection rate is low. We explore the use of deep neural networks to detect ChD from electrocardiograms (ECGs) to aid in the early detection of the disease.<h4>Methods</h4>We employ a convolutional neural network model that uses 12-lead ECG data to compute the probability of a ChD diagnosis. Our model is developed using two datasets which jointly comprise over two million entries from Brazilian patients: The SaMi-Trop study focusing on ChD patients, enriched with data from the CODE study from the general population. The model's performance is evaluated on two external datasets: the REDS-II, a study focused on ChD with 631 patients, and the ELSA-Brasil study, with 13,739 civil servant patients.<h4>Findings</h4>Evaluating our model, we obtain an AUC-ROC of 0.80 (CI 95% 0.79-0.82) for the validation set (samples from CODE and SaMi-Trop), and in external validation datasets: 0.68 (CI 95% 0.63-0.71) for REDS-II and 0.59 (CI 95% 0.56-0.63) for ELSA-Brasil. In the latter, we report a sensitivity of 0.52 (CI 95% 0.47-0.57) and 0.36 (CI 95% 0.30-0.42) and a specificity of 0.77 (CI 95% 0.72-0.81) and 0.76 (CI 95% 0.75-0.77), respectively. Additionally, when considering only patients with Chagas cardiomyopathy as positive, the model achieved an AUC-ROC of 0.82 (CI 95% 0.77-0.86) for REDS-II and 0.77 (CI 95% 0.68-0.85) for ELSA-Brasil.<h4>Interpretation</h4>The neural network detects chronic Chagas cardiomyopathy (CCC) from ECG-with weaker performance for early-stage cases. Future work should focus on curating large higher-quality datasets. The CODE dataset, our largest development dataset includes self-reported and therefore less reliable labels, limiting performance for non-CCC patients. Our findings can improve ChD detection and treatment, particularly in high-prevalence areas. |
| first_indexed | 2024-03-07T23:59:02Z |
| format | Article |
| id | doaj.art-a89cccbc52c948d68b9dbf04ae8810f2 |
| institution | Directory Open Access Journal |
| issn | 1935-2727 1935-2735 |
| language | English |
| last_indexed | 2024-03-07T23:59:02Z |
| publishDate | 2023-07-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS Neglected Tropical Diseases |
| spelling | doaj.art-a89cccbc52c948d68b9dbf04ae8810f22024-02-18T05:31:54ZengPublic Library of Science (PLoS)PLoS Neglected Tropical Diseases1935-27271935-27352023-07-01177e001111810.1371/journal.pntd.0011118Screening for Chagas disease from the electrocardiogram using a deep neural network.Carl JidlingDaniel GedonThomas B SchönClaudia Di Lorenzo OliveiraClareci Silva CardosoAriela Mota FerreiraLuana GiattiSandhi Maria BarretoEster C SabinoAntonio L P RibeiroAntônio H Ribeiro<h4>Background</h4>Worldwide, it is estimated that over 6 million people are infected with Chagas disease (ChD). It is a neglected disease that can lead to severe heart conditions in its chronic phase. While early treatment can avoid complications, the early-stage detection rate is low. We explore the use of deep neural networks to detect ChD from electrocardiograms (ECGs) to aid in the early detection of the disease.<h4>Methods</h4>We employ a convolutional neural network model that uses 12-lead ECG data to compute the probability of a ChD diagnosis. Our model is developed using two datasets which jointly comprise over two million entries from Brazilian patients: The SaMi-Trop study focusing on ChD patients, enriched with data from the CODE study from the general population. The model's performance is evaluated on two external datasets: the REDS-II, a study focused on ChD with 631 patients, and the ELSA-Brasil study, with 13,739 civil servant patients.<h4>Findings</h4>Evaluating our model, we obtain an AUC-ROC of 0.80 (CI 95% 0.79-0.82) for the validation set (samples from CODE and SaMi-Trop), and in external validation datasets: 0.68 (CI 95% 0.63-0.71) for REDS-II and 0.59 (CI 95% 0.56-0.63) for ELSA-Brasil. In the latter, we report a sensitivity of 0.52 (CI 95% 0.47-0.57) and 0.36 (CI 95% 0.30-0.42) and a specificity of 0.77 (CI 95% 0.72-0.81) and 0.76 (CI 95% 0.75-0.77), respectively. Additionally, when considering only patients with Chagas cardiomyopathy as positive, the model achieved an AUC-ROC of 0.82 (CI 95% 0.77-0.86) for REDS-II and 0.77 (CI 95% 0.68-0.85) for ELSA-Brasil.<h4>Interpretation</h4>The neural network detects chronic Chagas cardiomyopathy (CCC) from ECG-with weaker performance for early-stage cases. Future work should focus on curating large higher-quality datasets. The CODE dataset, our largest development dataset includes self-reported and therefore less reliable labels, limiting performance for non-CCC patients. Our findings can improve ChD detection and treatment, particularly in high-prevalence areas.https://journals.plos.org/plosntds/article/file?id=10.1371/journal.pntd.0011118&type=printable |
| spellingShingle | Carl Jidling Daniel Gedon Thomas B Schön Claudia Di Lorenzo Oliveira Clareci Silva Cardoso Ariela Mota Ferreira Luana Giatti Sandhi Maria Barreto Ester C Sabino Antonio L P Ribeiro Antônio H Ribeiro Screening for Chagas disease from the electrocardiogram using a deep neural network. PLoS Neglected Tropical Diseases |
| title | Screening for Chagas disease from the electrocardiogram using a deep neural network. |
| title_full | Screening for Chagas disease from the electrocardiogram using a deep neural network. |
| title_fullStr | Screening for Chagas disease from the electrocardiogram using a deep neural network. |
| title_full_unstemmed | Screening for Chagas disease from the electrocardiogram using a deep neural network. |
| title_short | Screening for Chagas disease from the electrocardiogram using a deep neural network. |
| title_sort | screening for chagas disease from the electrocardiogram using a deep neural network |
| url | https://journals.plos.org/plosntds/article/file?id=10.1371/journal.pntd.0011118&type=printable |
| work_keys_str_mv | AT carljidling screeningforchagasdiseasefromtheelectrocardiogramusingadeepneuralnetwork AT danielgedon screeningforchagasdiseasefromtheelectrocardiogramusingadeepneuralnetwork AT thomasbschon screeningforchagasdiseasefromtheelectrocardiogramusingadeepneuralnetwork AT claudiadilorenzooliveira screeningforchagasdiseasefromtheelectrocardiogramusingadeepneuralnetwork AT clarecisilvacardoso screeningforchagasdiseasefromtheelectrocardiogramusingadeepneuralnetwork AT arielamotaferreira screeningforchagasdiseasefromtheelectrocardiogramusingadeepneuralnetwork AT luanagiatti screeningforchagasdiseasefromtheelectrocardiogramusingadeepneuralnetwork AT sandhimariabarreto screeningforchagasdiseasefromtheelectrocardiogramusingadeepneuralnetwork AT estercsabino screeningforchagasdiseasefromtheelectrocardiogramusingadeepneuralnetwork AT antoniolpribeiro screeningforchagasdiseasefromtheelectrocardiogramusingadeepneuralnetwork AT antoniohribeiro screeningforchagasdiseasefromtheelectrocardiogramusingadeepneuralnetwork |