Analyzing breast cancer invasive disease event classification through explainable artificial intelligence
IntroductionRecently, accurate machine learning and deep learning approaches have been dedicated to the investigation of breast cancer invasive disease events (IDEs), such as recurrence, contralateral and second cancers. However, such approaches are poorly interpretable.MethodsThus, we designed an E...
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-02-01
|
| Series: | Frontiers in Medicine |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fmed.2023.1116354/full |
| _version_ | 1797935683616112640 |
|---|---|
| author | Raffaella Massafra Annarita Fanizzi Nicola Amoroso Nicola Amoroso Samantha Bove Maria Colomba Comes Domenico Pomarico Domenico Pomarico Vittorio Didonna Sergio Diotaiuti Luisa Galati Francesco Giotta Daniele La Forgia Agnese Latorre Angela Lombardi Annalisa Nardone Maria Irene Pastena Cosmo Maurizio Ressa Lucia Rinaldi Pasquale Tamborra Alfredo Zito Angelo Virgilio Paradiso Roberto Bellotti Roberto Bellotti Vito Lorusso |
| author_facet | Raffaella Massafra Annarita Fanizzi Nicola Amoroso Nicola Amoroso Samantha Bove Maria Colomba Comes Domenico Pomarico Domenico Pomarico Vittorio Didonna Sergio Diotaiuti Luisa Galati Francesco Giotta Daniele La Forgia Agnese Latorre Angela Lombardi Annalisa Nardone Maria Irene Pastena Cosmo Maurizio Ressa Lucia Rinaldi Pasquale Tamborra Alfredo Zito Angelo Virgilio Paradiso Roberto Bellotti Roberto Bellotti Vito Lorusso |
| author_sort | Raffaella Massafra |
| collection | DOAJ |
| description | IntroductionRecently, accurate machine learning and deep learning approaches have been dedicated to the investigation of breast cancer invasive disease events (IDEs), such as recurrence, contralateral and second cancers. However, such approaches are poorly interpretable.MethodsThus, we designed an Explainable Artificial Intelligence (XAI) framework to investigate IDEs within a cohort of 486 breast cancer patients enrolled at IRCCS Istituto Tumori “Giovanni Paolo II” in Bari, Italy. Using Shapley values, we determined the IDE driving features according to two periods, often adopted in clinical practice, of 5 and 10 years from the first tumor diagnosis.ResultsAge, tumor diameter, surgery type, and multiplicity are predominant within the 5-year frame, while therapy-related features, including hormone, chemotherapy schemes and lymphovascular invasion, dominate the 10-year IDE prediction. Estrogen Receptor (ER), proliferation marker Ki67 and metastatic lymph nodes affect both frames.DiscussionThus, our framework aims at shortening the distance between AI and clinical practice |
| first_indexed | 2024-04-10T18:19:05Z |
| format | Article |
| id | doaj.art-6a6b097c0dac4ec68f7e132a2023c0ef |
| institution | Directory Open Access Journal |
| issn | 2296-858X |
| language | English |
| last_indexed | 2024-04-10T18:19:05Z |
| publishDate | 2023-02-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Medicine |
| spelling | doaj.art-6a6b097c0dac4ec68f7e132a2023c0ef2023-02-02T07:55:06ZengFrontiers Media S.A.Frontiers in Medicine2296-858X2023-02-011010.3389/fmed.2023.11163541116354Analyzing breast cancer invasive disease event classification through explainable artificial intelligenceRaffaella Massafra0Annarita Fanizzi1Nicola Amoroso2Nicola Amoroso3Samantha Bove4Maria Colomba Comes5Domenico Pomarico6Domenico Pomarico7Vittorio Didonna8Sergio Diotaiuti9Luisa Galati10Francesco Giotta11Daniele La Forgia12Agnese Latorre13Angela Lombardi14Annalisa Nardone15Maria Irene Pastena16Cosmo Maurizio Ressa17Lucia Rinaldi18Pasquale Tamborra19Alfredo Zito20Angelo Virgilio Paradiso21Roberto Bellotti22Roberto Bellotti23Vito Lorusso24IRCCS Istituto Tumori “Giovanni Paolo II”, Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II”, Bari, ItalyINFN, Sezione di Bari, Bari, ItalyDipartimento di Farmacia-Scienze del Farmaco, Università degli Studi di Bari Aldo Moro, Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II”, Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II”, Bari, ItalyINFN, Sezione di Bari, Bari, ItalyDipartimento di Fisica, Università degli Studi di Bari Aldo Moro, Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II”, Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II”, Bari, ItalyInternational Agency for Research on Cancer, Lyon, FranceIRCCS Istituto Tumori “Giovanni Paolo II”, Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II”, Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II”, Bari, ItalyDipartimento di Ingegneria Elettrica e dell'Informazione, Politecnico di Bari, Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II”, Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II”, Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II”, Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II”, Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II”, Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II”, Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II”, Bari, ItalyINFN, Sezione di Bari, Bari, ItalyDipartimento di Fisica, Università degli Studi di Bari Aldo Moro, Bari, ItalyIRCCS Istituto Tumori “Giovanni Paolo II”, Bari, ItalyIntroductionRecently, accurate machine learning and deep learning approaches have been dedicated to the investigation of breast cancer invasive disease events (IDEs), such as recurrence, contralateral and second cancers. However, such approaches are poorly interpretable.MethodsThus, we designed an Explainable Artificial Intelligence (XAI) framework to investigate IDEs within a cohort of 486 breast cancer patients enrolled at IRCCS Istituto Tumori “Giovanni Paolo II” in Bari, Italy. Using Shapley values, we determined the IDE driving features according to two periods, often adopted in clinical practice, of 5 and 10 years from the first tumor diagnosis.ResultsAge, tumor diameter, surgery type, and multiplicity are predominant within the 5-year frame, while therapy-related features, including hormone, chemotherapy schemes and lymphovascular invasion, dominate the 10-year IDE prediction. Estrogen Receptor (ER), proliferation marker Ki67 and metastatic lymph nodes affect both frames.DiscussionThus, our framework aims at shortening the distance between AI and clinical practicehttps://www.frontiersin.org/articles/10.3389/fmed.2023.1116354/fullinvasive disease eventsbreast cancerexplainable AI10-year follow up5-year follow up |
| spellingShingle | Raffaella Massafra Annarita Fanizzi Nicola Amoroso Nicola Amoroso Samantha Bove Maria Colomba Comes Domenico Pomarico Domenico Pomarico Vittorio Didonna Sergio Diotaiuti Luisa Galati Francesco Giotta Daniele La Forgia Agnese Latorre Angela Lombardi Annalisa Nardone Maria Irene Pastena Cosmo Maurizio Ressa Lucia Rinaldi Pasquale Tamborra Alfredo Zito Angelo Virgilio Paradiso Roberto Bellotti Roberto Bellotti Vito Lorusso Analyzing breast cancer invasive disease event classification through explainable artificial intelligence Frontiers in Medicine invasive disease events breast cancer explainable AI 10-year follow up 5-year follow up |
| title | Analyzing breast cancer invasive disease event classification through explainable artificial intelligence |
| title_full | Analyzing breast cancer invasive disease event classification through explainable artificial intelligence |
| title_fullStr | Analyzing breast cancer invasive disease event classification through explainable artificial intelligence |
| title_full_unstemmed | Analyzing breast cancer invasive disease event classification through explainable artificial intelligence |
| title_short | Analyzing breast cancer invasive disease event classification through explainable artificial intelligence |
| title_sort | analyzing breast cancer invasive disease event classification through explainable artificial intelligence |
| topic | invasive disease events breast cancer explainable AI 10-year follow up 5-year follow up |
| url | https://www.frontiersin.org/articles/10.3389/fmed.2023.1116354/full |
| work_keys_str_mv | AT raffaellamassafra analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT annaritafanizzi analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT nicolaamoroso analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT nicolaamoroso analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT samanthabove analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT mariacolombacomes analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT domenicopomarico analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT domenicopomarico analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT vittoriodidonna analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT sergiodiotaiuti analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT luisagalati analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT francescogiotta analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT danielelaforgia analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT agneselatorre analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT angelalombardi analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT annalisanardone analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT mariairenepastena analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT cosmomaurizioressa analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT luciarinaldi analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT pasqualetamborra analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT alfredozito analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT angelovirgilioparadiso analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT robertobellotti analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT robertobellotti analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence AT vitolorusso analyzingbreastcancerinvasivediseaseeventclassificationthroughexplainableartificialintelligence |