UMAP Based Anomaly Detection for Minimal Residual Disease Quantification within Acute Myeloid Leukemia
Leukemia is the most frequent malignancy in children and adolescents, with acute lymphoblastic leukemia (ALL) and acute myeloid leukemia (AML) as the most common subtypes. Minimal residual disease (MRD) measured by flow cytometry (FCM) has proven to be a strong prognostic factor in ALL as well as in...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-02-01
|
| Series: | Cancers |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-6694/14/4/898 |
| _version_ | 1797482108027928576 |
|---|---|
| author | Lisa Weijler Florian Kowarsch Matthias Wödlinger Michael Reiter Margarita Maurer-Granofszky Angela Schumich Michael N. Dworzak |
| author_facet | Lisa Weijler Florian Kowarsch Matthias Wödlinger Michael Reiter Margarita Maurer-Granofszky Angela Schumich Michael N. Dworzak |
| author_sort | Lisa Weijler |
| collection | DOAJ |
| description | Leukemia is the most frequent malignancy in children and adolescents, with acute lymphoblastic leukemia (ALL) and acute myeloid leukemia (AML) as the most common subtypes. Minimal residual disease (MRD) measured by flow cytometry (FCM) has proven to be a strong prognostic factor in ALL as well as in AML. Machine learning techniques have been emerging in the field of automated MRD quantification with the objective of superseding subjective and time-consuming manual analysis of FCM-MRD data. In contrast to ALL, where supervised multi-class classification methods have been successfully deployed for MRD detection, AML poses new challenges: AML is rarer (with fewer available training data) than ALL and much more heterogeneous in its immunophenotypic appearance, where one-class classification (anomaly detection) methods seem more suitable. In this work, a new semi-supervised approach based on the UMAP algorithm for MRD detection utilizing only labels of blast free FCM samples is presented. The method is tested on a newly gathered set of AML FCM samples and results are compared to state-of-the-art methods. We reach a median <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>F</mi><mn>1</mn></msub></semantics></math></inline-formula>-score of 0.794, while providing a transparent classification pipeline with explainable results that facilitates inter-disciplinary work between medical and technical experts. This work shows that despite several issues yet to overcome, the merits of automated MRD quantification can be fully exploited also in AML. |
| first_indexed | 2024-03-09T22:24:35Z |
| format | Article |
| id | doaj.art-e538688cc8c1415cafa4db95f1f7aa32 |
| institution | Directory Open Access Journal |
| issn | 2072-6694 |
| language | English |
| last_indexed | 2024-03-09T22:24:35Z |
| publishDate | 2022-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Cancers |
| spelling | doaj.art-e538688cc8c1415cafa4db95f1f7aa322023-11-23T19:08:09ZengMDPI AGCancers2072-66942022-02-0114489810.3390/cancers14040898UMAP Based Anomaly Detection for Minimal Residual Disease Quantification within Acute Myeloid LeukemiaLisa Weijler0Florian Kowarsch1Matthias Wödlinger2Michael Reiter3Margarita Maurer-Granofszky4Angela Schumich5Michael N. Dworzak6Computer Vision Lab, Faculty of Informatics, Technical University of Vienna, 1040 Vienna, AustriaComputer Vision Lab, Faculty of Informatics, Technical University of Vienna, 1040 Vienna, AustriaComputer Vision Lab, Faculty of Informatics, Technical University of Vienna, 1040 Vienna, AustriaComputer Vision Lab, Faculty of Informatics, Technical University of Vienna, 1040 Vienna, AustriaImmunological Diagnostics, St. Anna Children’s Cancer Research Institute (CCRI), 1090 Vienna, AustriaImmunological Diagnostics, St. Anna Children’s Cancer Research Institute (CCRI), 1090 Vienna, AustriaImmunological Diagnostics, St. Anna Children’s Cancer Research Institute (CCRI), 1090 Vienna, AustriaLeukemia is the most frequent malignancy in children and adolescents, with acute lymphoblastic leukemia (ALL) and acute myeloid leukemia (AML) as the most common subtypes. Minimal residual disease (MRD) measured by flow cytometry (FCM) has proven to be a strong prognostic factor in ALL as well as in AML. Machine learning techniques have been emerging in the field of automated MRD quantification with the objective of superseding subjective and time-consuming manual analysis of FCM-MRD data. In contrast to ALL, where supervised multi-class classification methods have been successfully deployed for MRD detection, AML poses new challenges: AML is rarer (with fewer available training data) than ALL and much more heterogeneous in its immunophenotypic appearance, where one-class classification (anomaly detection) methods seem more suitable. In this work, a new semi-supervised approach based on the UMAP algorithm for MRD detection utilizing only labels of blast free FCM samples is presented. The method is tested on a newly gathered set of AML FCM samples and results are compared to state-of-the-art methods. We reach a median <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>F</mi><mn>1</mn></msub></semantics></math></inline-formula>-score of 0.794, while providing a transparent classification pipeline with explainable results that facilitates inter-disciplinary work between medical and technical experts. This work shows that despite several issues yet to overcome, the merits of automated MRD quantification can be fully exploited also in AML.https://www.mdpi.com/2072-6694/14/4/898acute myeloid leukemiaanomaly detectionUMAPset-transformerself-attentionflow cytometry |
| spellingShingle | Lisa Weijler Florian Kowarsch Matthias Wödlinger Michael Reiter Margarita Maurer-Granofszky Angela Schumich Michael N. Dworzak UMAP Based Anomaly Detection for Minimal Residual Disease Quantification within Acute Myeloid Leukemia Cancers acute myeloid leukemia anomaly detection UMAP set-transformer self-attention flow cytometry |
| title | UMAP Based Anomaly Detection for Minimal Residual Disease Quantification within Acute Myeloid Leukemia |
| title_full | UMAP Based Anomaly Detection for Minimal Residual Disease Quantification within Acute Myeloid Leukemia |
| title_fullStr | UMAP Based Anomaly Detection for Minimal Residual Disease Quantification within Acute Myeloid Leukemia |
| title_full_unstemmed | UMAP Based Anomaly Detection for Minimal Residual Disease Quantification within Acute Myeloid Leukemia |
| title_short | UMAP Based Anomaly Detection for Minimal Residual Disease Quantification within Acute Myeloid Leukemia |
| title_sort | umap based anomaly detection for minimal residual disease quantification within acute myeloid leukemia |
| topic | acute myeloid leukemia anomaly detection UMAP set-transformer self-attention flow cytometry |
| url | https://www.mdpi.com/2072-6694/14/4/898 |
| work_keys_str_mv | AT lisaweijler umapbasedanomalydetectionforminimalresidualdiseasequantificationwithinacutemyeloidleukemia AT floriankowarsch umapbasedanomalydetectionforminimalresidualdiseasequantificationwithinacutemyeloidleukemia AT matthiaswodlinger umapbasedanomalydetectionforminimalresidualdiseasequantificationwithinacutemyeloidleukemia AT michaelreiter umapbasedanomalydetectionforminimalresidualdiseasequantificationwithinacutemyeloidleukemia AT margaritamaurergranofszky umapbasedanomalydetectionforminimalresidualdiseasequantificationwithinacutemyeloidleukemia AT angelaschumich umapbasedanomalydetectionforminimalresidualdiseasequantificationwithinacutemyeloidleukemia AT michaelndworzak umapbasedanomalydetectionforminimalresidualdiseasequantificationwithinacutemyeloidleukemia |