{"id":{"repo_id":"cornell","oai_identifier":"oai:ecommons.cornell.edu:1813/111653"},"canonical_url":"https://search.dev.ndltd.org/etd/cornell/oai:ecommons.cornell.edu:1813/111653","repository":{"repo_id":"cornell","name":"Cornell University","base_url":"https://ecommons.cornell.edu/server/oai/request"},"display":{"title":"SYNTHETIC OVERSAMPLING IMPROVES SPECTRAL DETECTION OF AFLATOXIN IN SINGLE MAIZE KERNELS","abstract":"Current spectral models for detection of aflatoxin (AF) in maize kernels are limited by very low representation of kernels with AF levels above the models’ specified contamination thresholds. With the aim of improving prediction model sensitivity to AF-contaminated kernels, two methods were evaluated for artificially enriching the representation of kernels containing ≥150 ppb AF in a large spectral dataset: humid incubation and synthetic oversampling. The addition of synthetic training samples improved prediction accuracy of kernels containing ≥150 ppb AF from 51% to 80%. Humid incubation contributed additional kernels with intermediate levels of AF contamination (primarily 5 - 75 ppb), but generally did not change the overall distribution of AF contamination. Feature importance distributions overlapped among models at 329-345 nm, 380-385.5 nm, 415-425 nm, 639-668 nm, and 1,013.5-1,060 nm. These spectral ranges could be applicable to the development of limited wavelength grain sorting devices for low-resource settings.","abstract_html":"Current spectral models for detection of aflatoxin (AF) in maize kernels are limited by very low representation of kernels with AF levels above the models’ specified contamination thresholds. With the aim of improving prediction model sensitivity to AF-contaminated kernels, two methods were evaluated for artificially enriching the representation of kernels containing ≥150 ppb AF in a large spectral dataset: humid incubation and synthetic oversampling. The addition of synthetic training samples improved prediction accuracy of kernels containing ≥150 ppb AF from 51% to 80%. Humid incubation contributed additional kernels with intermediate levels of AF contamination (primarily 5 - 75 ppb), but generally did not change the overall distribution of AF contamination. Feature importance distributions overlapped among models at 329-345 nm, 380-385.5 nm, 415-425 nm, 639-668 nm, and 1,013.5-1,060 nm. These spectral ranges could be applicable to the development of limited wavelength grain sorting devices for low-resource settings.","abstract_has_math":false,"creators":["Siegel, Chloe Siobhan"],"institution":"Cornell University","degree_name":"M.S., Plant Pathology and Plant-Microbe Biology","degree_level":"Master of Science","degree_discipline":"Plant Pathology and Plant-Microbe Biology","degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":["Bergstrom, Gary Carlton"],"year":2022,"date_issued":"2022-05","date_published":"2022-05","updated_at":"2026-07-24T01:48:56Z","subjects":[],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier.doi","label":"DOI","values":["https://doi.org/10.7298/69xa-vr67"],"render_values":[{"text":"https://doi.org/10.7298/69xa-vr67","href":"https://doi.org/10.7298/69xa-vr67","code":true}]},{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["ProQuest Submission ID: 11459","ProQuest Publication ID: 29168023"],"render_values":[{"text":"ProQuest Submission ID: 11459","href":null,"code":true},{"text":"ProQuest Publication ID: 29168023","href":null,"code":true}]}]},"links":{"outbound_url":"https://hdl.handle.net/1813/111653","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Bergstrom, Gary Carlton"]},{"key":"dc:creator","label":"Author","values":["Siegel, Chloe Siobhan"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2022-09-15T15:49:20Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2022-09-15T15:49:20Z"]},{"key":"dc:date.issued","label":"Date","values":["2022-05"]},{"key":"dc:type","label":"Dc Type","values":["dissertation or thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Plant Pathology and Plant-Microbe Biology"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Master of Science"]},{"key":"thesis:degree_name","label":"Degree Name","values":["M.S., Plant Pathology and Plant-Microbe Biology"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["Cornell University"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.doi","label":"DOI","values":["https://doi.org/10.7298/69xa-vr67"]},{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["ProQuest Submission ID: 11459","ProQuest Publication ID: 29168023"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1813/111653"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["131 pages"]},{"key":"dc:description.abstract","label":"Abstract","values":["Current spectral models for detection of aflatoxin (AF) in maize kernels are limited by very low representation of kernels with AF levels above the models’ specified contamination thresholds. With the aim of improving prediction model sensitivity to AF-contaminated kernels, two methods were evaluated for artificially enriching the representation of kernels containing ≥150 ppb AF in a large spectral dataset: humid incubation and synthetic oversampling. The addition of synthetic training samples improved prediction accuracy of kernels containing ≥150 ppb AF from 51% to 80%. Humid incubation contributed additional kernels with intermediate levels of AF contamination (primarily 5 - 75 ppb), but generally did not change the overall distribution of AF contamination. Feature importance distributions overlapped among models at 329-345 nm, 380-385.5 nm, 415-425 nm, 639-668 nm, and 1,013.5-1,060 nm. These spectral ranges could be applicable to the development of limited wavelength grain sorting devices for low-resource settings."]},{"key":"dc:format.mimetype","label":"Dc Format Mimetype","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["SYNTHETIC OVERSAMPLING IMPROVES SPECTRAL DETECTION OF AFLATOXIN IN SINGLE MAIZE KERNELS"]}]}],"canonical_facts":{"dc:contributor.committeemember":["Bergstrom, Gary Carlton"],"dc:creator":["Siegel, Chloe Siobhan"],"dc:date.accessioned":["2022-09-15T15:49:20Z"],"dc:date.available":["2022-09-15T15:49:20Z"],"dc:date.issued":["2022-05"],"dc:description":["131 pages"],"dc:description.abstract":["Current spectral models for detection of aflatoxin (AF) in maize kernels are limited by very low representation of kernels with AF levels above the models’ specified contamination thresholds. With the aim of improving prediction model sensitivity to AF-contaminated kernels, two methods were evaluated for artificially enriching the representation of kernels containing ≥150 ppb AF in a large spectral dataset: humid incubation and synthetic oversampling. The addition of synthetic training samples improved prediction accuracy of kernels containing ≥150 ppb AF from 51% to 80%. Humid incubation contributed additional kernels with intermediate levels of AF contamination (primarily 5 - 75 ppb), but generally did not change the overall distribution of AF contamination. Feature importance distributions overlapped among models at 329-345 nm, 380-385.5 nm, 415-425 nm, 639-668 nm, and 1,013.5-1,060 nm. These spectral ranges could be applicable to the development of limited wavelength grain sorting devices for low-resource settings."],"dc:format.mimetype":["application/pdf"],"dc:identifier.doi":["https://doi.org/10.7298/69xa-vr67"],"dc:identifier.other":["ProQuest Submission ID: 11459","ProQuest Publication ID: 29168023"],"dc:identifier.uri":["https://hdl.handle.net/1813/111653"],"dc:language.iso":["en"],"dc:title":["SYNTHETIC OVERSAMPLING IMPROVES SPECTRAL DETECTION OF AFLATOXIN IN SINGLE MAIZE KERNELS"],"dc:type":["dissertation or thesis"],"thesis:degree_discipline":["Plant Pathology and Plant-Microbe Biology"],"thesis:degree_level":["Master of Science"],"thesis:degree_name":["M.S., Plant Pathology and Plant-Microbe Biology"],"thesis:institution_name":["Cornell University"]},"updated_at":"2026-07-24T01:48:56Z"}