Explainable Deep Learning Reproduces a ‘Professional Eye’ on the Diagnosis of Internal Disorders in Persimmon Fruit

Abstract Recent rapid progress in deep neural network techniques has allowed recognition and classification of various objects, often exceeding the performance of the human eye. In plant biology and crop sciences, some deep neural network frameworks have been applied mainly for effective and rapid p...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Plant and cell physiology 2020-12, Vol.61 (11), p.1967-1973
Hauptverfasser: Akagi, Takashi, Onishi, Masanori, Masuda, Kanae, Kuroki, Ryohei, Baba, Kohei, Takeshita, Kouki, Suzuki, Tetsuya, Niikawa, Takeshi, Uchida, Seiichi, Ise, Takeshi
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page 1973
container_issue 11
container_start_page 1967
container_title Plant and cell physiology
container_volume 61
creator Akagi, Takashi
Onishi, Masanori
Masuda, Kanae
Kuroki, Ryohei
Baba, Kohei
Takeshita, Kouki
Suzuki, Tetsuya
Niikawa, Takeshi
Uchida, Seiichi
Ise, Takeshi
description Abstract Recent rapid progress in deep neural network techniques has allowed recognition and classification of various objects, often exceeding the performance of the human eye. In plant biology and crop sciences, some deep neural network frameworks have been applied mainly for effective and rapid phenotyping. In this study, beyond simple optimizations of phenotyping, we propose an application of deep neural networks to make an image-based internal disorder diagnosis that is hard even for experts, and to visualize the reasons behind each diagnosis to provide biological interpretations. Here, we exemplified classification of calyx-end cracking in persimmon fruit by using five convolutional neural network models with various layer structures and examined potential analytical options involved in the diagnostic qualities. With 3,173 visible RGB images from the fruit apex side, the neural networks successfully made the binary classification of each degree of disorder, with up to 90% accuracy. Furthermore, feature visualizations, such as Grad-CAM and LRP, visualize the regions of the image that contribute to the diagnosis. They suggest that specific patterns of color unevenness, such as in the fruit peripheral area, can be indexes of calyx-end cracking. These results not only provided novel insights into indexes of fruit internal disorders but also proposed the potential applicability of deep neural networks in plant biology.
doi_str_mv 10.1093/pcp/pcaa111
format Article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_2437398858</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><oup_id>10.1093/pcp/pcaa111</oup_id><sourcerecordid>2437398858</sourcerecordid><originalsourceid>FETCH-LOGICAL-c423t-ac0076856d96c6672f36009811b559e96d8db1083c0da06e47e5c41341b747163</originalsourceid><addsrcrecordid>eNp9kDFPwzAQhS0EolCY2JEnhIQKdp04zoighUqVqBDMkeNcilFiBzsRdOvPgL_XX4KrFsTEcLobvvd07yF0QsklJSm7alQTRkpK6Q46oFFCBymJ2e6fu4cOvX8lhCSCkX3UY0MRxYwkB-h99NFUUhuZV4BvARo8BemMNnP8CI2zRafAY4lXy8-ZsyV4r62RFR4tYLX8wtbg9iUItZwb67XHtsQT04JbM7faW1eA81gbPAtb13UQjF2n2yO0V8rKw_F299HzePR0cz-YPtxNbq6nAxUNWTuQKvzMRcyLlCvOk2HJOCGpoDSP4xRSXogip0QwRQpJOEQJxCqiLKJ5EsJz1kfnG9-Q5a0D32a19gqqShqwnc-GEUtYKkQsAnqxQZWz3jsos8bpWrpFRkm2bjoLTWfbpgN9ujXu8hqKX_an2gCcbQDbNf86fQP6s4kY</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2437398858</pqid></control><display><type>article</type><title>Explainable Deep Learning Reproduces a ‘Professional Eye’ on the Diagnosis of Internal Disorders in Persimmon Fruit</title><source>MEDLINE</source><source>Oxford University Press Journals All Titles (1996-Current)</source><source>EZB-FREE-00999 freely available EZB journals</source><source>Alma/SFX Local Collection</source><creator>Akagi, Takashi ; Onishi, Masanori ; Masuda, Kanae ; Kuroki, Ryohei ; Baba, Kohei ; Takeshita, Kouki ; Suzuki, Tetsuya ; Niikawa, Takeshi ; Uchida, Seiichi ; Ise, Takeshi</creator><creatorcontrib>Akagi, Takashi ; Onishi, Masanori ; Masuda, Kanae ; Kuroki, Ryohei ; Baba, Kohei ; Takeshita, Kouki ; Suzuki, Tetsuya ; Niikawa, Takeshi ; Uchida, Seiichi ; Ise, Takeshi</creatorcontrib><description>Abstract Recent rapid progress in deep neural network techniques has allowed recognition and classification of various objects, often exceeding the performance of the human eye. In plant biology and crop sciences, some deep neural network frameworks have been applied mainly for effective and rapid phenotyping. In this study, beyond simple optimizations of phenotyping, we propose an application of deep neural networks to make an image-based internal disorder diagnosis that is hard even for experts, and to visualize the reasons behind each diagnosis to provide biological interpretations. Here, we exemplified classification of calyx-end cracking in persimmon fruit by using five convolutional neural network models with various layer structures and examined potential analytical options involved in the diagnostic qualities. With 3,173 visible RGB images from the fruit apex side, the neural networks successfully made the binary classification of each degree of disorder, with up to 90% accuracy. Furthermore, feature visualizations, such as Grad-CAM and LRP, visualize the regions of the image that contribute to the diagnosis. They suggest that specific patterns of color unevenness, such as in the fruit peripheral area, can be indexes of calyx-end cracking. These results not only provided novel insights into indexes of fruit internal disorders but also proposed the potential applicability of deep neural networks in plant biology.</description><identifier>ISSN: 1471-9053</identifier><identifier>EISSN: 1471-9053</identifier><identifier>DOI: 10.1093/pcp/pcaa111</identifier><identifier>PMID: 32845307</identifier><language>eng</language><publisher>Japan: Oxford University Press</publisher><subject>Deep Learning ; Diospyros - anatomy &amp; histology ; Flowers - anatomy &amp; histology ; Fruit - anatomy &amp; histology ; Image Interpretation, Computer-Assisted ; Neural Networks, Computer ; Plant Diseases</subject><ispartof>Plant and cell physiology, 2020-12, Vol.61 (11), p.1967-1973</ispartof><rights>The Author(s) 2020. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists. All rights reserved. For permissions, please email: journals.permissions@oup.com 2020</rights><rights>The Author(s) 2020. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists. All rights reserved. For permissions, please email: journals.permissions@oup.com.</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c423t-ac0076856d96c6672f36009811b559e96d8db1083c0da06e47e5c41341b747163</citedby><cites>FETCH-LOGICAL-c423t-ac0076856d96c6672f36009811b559e96d8db1083c0da06e47e5c41341b747163</cites><orcidid>0000-0001-9993-8880</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,780,784,1584,27924,27925</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/32845307$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Akagi, Takashi</creatorcontrib><creatorcontrib>Onishi, Masanori</creatorcontrib><creatorcontrib>Masuda, Kanae</creatorcontrib><creatorcontrib>Kuroki, Ryohei</creatorcontrib><creatorcontrib>Baba, Kohei</creatorcontrib><creatorcontrib>Takeshita, Kouki</creatorcontrib><creatorcontrib>Suzuki, Tetsuya</creatorcontrib><creatorcontrib>Niikawa, Takeshi</creatorcontrib><creatorcontrib>Uchida, Seiichi</creatorcontrib><creatorcontrib>Ise, Takeshi</creatorcontrib><title>Explainable Deep Learning Reproduces a ‘Professional Eye’ on the Diagnosis of Internal Disorders in Persimmon Fruit</title><title>Plant and cell physiology</title><addtitle>Plant Cell Physiol</addtitle><description>Abstract Recent rapid progress in deep neural network techniques has allowed recognition and classification of various objects, often exceeding the performance of the human eye. In plant biology and crop sciences, some deep neural network frameworks have been applied mainly for effective and rapid phenotyping. In this study, beyond simple optimizations of phenotyping, we propose an application of deep neural networks to make an image-based internal disorder diagnosis that is hard even for experts, and to visualize the reasons behind each diagnosis to provide biological interpretations. Here, we exemplified classification of calyx-end cracking in persimmon fruit by using five convolutional neural network models with various layer structures and examined potential analytical options involved in the diagnostic qualities. With 3,173 visible RGB images from the fruit apex side, the neural networks successfully made the binary classification of each degree of disorder, with up to 90% accuracy. Furthermore, feature visualizations, such as Grad-CAM and LRP, visualize the regions of the image that contribute to the diagnosis. They suggest that specific patterns of color unevenness, such as in the fruit peripheral area, can be indexes of calyx-end cracking. These results not only provided novel insights into indexes of fruit internal disorders but also proposed the potential applicability of deep neural networks in plant biology.</description><subject>Deep Learning</subject><subject>Diospyros - anatomy &amp; histology</subject><subject>Flowers - anatomy &amp; histology</subject><subject>Fruit - anatomy &amp; histology</subject><subject>Image Interpretation, Computer-Assisted</subject><subject>Neural Networks, Computer</subject><subject>Plant Diseases</subject><issn>1471-9053</issn><issn>1471-9053</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>EIF</sourceid><recordid>eNp9kDFPwzAQhS0EolCY2JEnhIQKdp04zoighUqVqBDMkeNcilFiBzsRdOvPgL_XX4KrFsTEcLobvvd07yF0QsklJSm7alQTRkpK6Q46oFFCBymJ2e6fu4cOvX8lhCSCkX3UY0MRxYwkB-h99NFUUhuZV4BvARo8BemMNnP8CI2zRafAY4lXy8-ZsyV4r62RFR4tYLX8wtbg9iUItZwb67XHtsQT04JbM7faW1eA81gbPAtb13UQjF2n2yO0V8rKw_F299HzePR0cz-YPtxNbq6nAxUNWTuQKvzMRcyLlCvOk2HJOCGpoDSP4xRSXogip0QwRQpJOEQJxCqiLKJ5EsJz1kfnG9-Q5a0D32a19gqqShqwnc-GEUtYKkQsAnqxQZWz3jsos8bpWrpFRkm2bjoLTWfbpgN9ujXu8hqKX_an2gCcbQDbNf86fQP6s4kY</recordid><startdate>20201223</startdate><enddate>20201223</enddate><creator>Akagi, Takashi</creator><creator>Onishi, Masanori</creator><creator>Masuda, Kanae</creator><creator>Kuroki, Ryohei</creator><creator>Baba, Kohei</creator><creator>Takeshita, Kouki</creator><creator>Suzuki, Tetsuya</creator><creator>Niikawa, Takeshi</creator><creator>Uchida, Seiichi</creator><creator>Ise, Takeshi</creator><general>Oxford University Press</general><scope>CGR</scope><scope>CUY</scope><scope>CVF</scope><scope>ECM</scope><scope>EIF</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><orcidid>https://orcid.org/0000-0001-9993-8880</orcidid></search><sort><creationdate>20201223</creationdate><title>Explainable Deep Learning Reproduces a ‘Professional Eye’ on the Diagnosis of Internal Disorders in Persimmon Fruit</title><author>Akagi, Takashi ; Onishi, Masanori ; Masuda, Kanae ; Kuroki, Ryohei ; Baba, Kohei ; Takeshita, Kouki ; Suzuki, Tetsuya ; Niikawa, Takeshi ; Uchida, Seiichi ; Ise, Takeshi</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c423t-ac0076856d96c6672f36009811b559e96d8db1083c0da06e47e5c41341b747163</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Deep Learning</topic><topic>Diospyros - anatomy &amp; histology</topic><topic>Flowers - anatomy &amp; histology</topic><topic>Fruit - anatomy &amp; histology</topic><topic>Image Interpretation, Computer-Assisted</topic><topic>Neural Networks, Computer</topic><topic>Plant Diseases</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Akagi, Takashi</creatorcontrib><creatorcontrib>Onishi, Masanori</creatorcontrib><creatorcontrib>Masuda, Kanae</creatorcontrib><creatorcontrib>Kuroki, Ryohei</creatorcontrib><creatorcontrib>Baba, Kohei</creatorcontrib><creatorcontrib>Takeshita, Kouki</creatorcontrib><creatorcontrib>Suzuki, Tetsuya</creatorcontrib><creatorcontrib>Niikawa, Takeshi</creatorcontrib><creatorcontrib>Uchida, Seiichi</creatorcontrib><creatorcontrib>Ise, Takeshi</creatorcontrib><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><jtitle>Plant and cell physiology</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Akagi, Takashi</au><au>Onishi, Masanori</au><au>Masuda, Kanae</au><au>Kuroki, Ryohei</au><au>Baba, Kohei</au><au>Takeshita, Kouki</au><au>Suzuki, Tetsuya</au><au>Niikawa, Takeshi</au><au>Uchida, Seiichi</au><au>Ise, Takeshi</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Explainable Deep Learning Reproduces a ‘Professional Eye’ on the Diagnosis of Internal Disorders in Persimmon Fruit</atitle><jtitle>Plant and cell physiology</jtitle><addtitle>Plant Cell Physiol</addtitle><date>2020-12-23</date><risdate>2020</risdate><volume>61</volume><issue>11</issue><spage>1967</spage><epage>1973</epage><pages>1967-1973</pages><issn>1471-9053</issn><eissn>1471-9053</eissn><abstract>Abstract Recent rapid progress in deep neural network techniques has allowed recognition and classification of various objects, often exceeding the performance of the human eye. In plant biology and crop sciences, some deep neural network frameworks have been applied mainly for effective and rapid phenotyping. In this study, beyond simple optimizations of phenotyping, we propose an application of deep neural networks to make an image-based internal disorder diagnosis that is hard even for experts, and to visualize the reasons behind each diagnosis to provide biological interpretations. Here, we exemplified classification of calyx-end cracking in persimmon fruit by using five convolutional neural network models with various layer structures and examined potential analytical options involved in the diagnostic qualities. With 3,173 visible RGB images from the fruit apex side, the neural networks successfully made the binary classification of each degree of disorder, with up to 90% accuracy. Furthermore, feature visualizations, such as Grad-CAM and LRP, visualize the regions of the image that contribute to the diagnosis. They suggest that specific patterns of color unevenness, such as in the fruit peripheral area, can be indexes of calyx-end cracking. These results not only provided novel insights into indexes of fruit internal disorders but also proposed the potential applicability of deep neural networks in plant biology.</abstract><cop>Japan</cop><pub>Oxford University Press</pub><pmid>32845307</pmid><doi>10.1093/pcp/pcaa111</doi><tpages>7</tpages><orcidid>https://orcid.org/0000-0001-9993-8880</orcidid><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 1471-9053
ispartof Plant and cell physiology, 2020-12, Vol.61 (11), p.1967-1973
issn 1471-9053
1471-9053
language eng
recordid cdi_proquest_miscellaneous_2437398858
source MEDLINE; Oxford University Press Journals All Titles (1996-Current); EZB-FREE-00999 freely available EZB journals; Alma/SFX Local Collection
subjects Deep Learning
Diospyros - anatomy & histology
Flowers - anatomy & histology
Fruit - anatomy & histology
Image Interpretation, Computer-Assisted
Neural Networks, Computer
Plant Diseases
title Explainable Deep Learning Reproduces a ‘Professional Eye’ on the Diagnosis of Internal Disorders in Persimmon Fruit
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-02T20%3A05%3A15IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Explainable%20Deep%20Learning%20Reproduces%20a%20%E2%80%98Professional%20Eye%E2%80%99%20on%20the%20Diagnosis%20of%20Internal%20Disorders%20in%20Persimmon%20Fruit&rft.jtitle=Plant%20and%20cell%20physiology&rft.au=Akagi,%20Takashi&rft.date=2020-12-23&rft.volume=61&rft.issue=11&rft.spage=1967&rft.epage=1973&rft.pages=1967-1973&rft.issn=1471-9053&rft.eissn=1471-9053&rft_id=info:doi/10.1093/pcp/pcaa111&rft_dat=%3Cproquest_cross%3E2437398858%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2437398858&rft_id=info:pmid/32845307&rft_oup_id=10.1093/pcp/pcaa111&rfr_iscdi=true