{"corpus_id":155537410,"paper_sha":"17a227c6100b6add4069a7e559763c49d5e7d989","doi":"10.1109/ISCAS.2019.8702211","arxiv_id":null,"pmid":null,"pmcid":null,"mag_id":2942968344,"dblp_id":"conf/iscas/ZhanGK19","acl_id":null,"title":"A Resource-Optimized VLSI Architecture for Patient-Specific Seizure Detection using Frontal-Lobe EEG","year":2019,"publication_date":"2019-05-01","venue":"International Symposium on Circuits and Systems","journal":{"name":"2019 IEEE International Symposium on Circuits and Systems (ISCAS)","pages":"1-5","volume":null},"journal_issn":null,"journal_title":null,"publication_types":["JournalArticle"],"pubmed_pub_types":null,"s2_fields_of_study":["Medicine","Computer Science","Engineering"],"reference_count":20,"citation_count":13,"influential_citation_count":1,"is_open_access":false,"arxiv_categories":null,"arxiv_license":null,"arxiv_journal_ref":null,"mesh_headings":null,"chemicals":null,"comments_corrections":null,"source_flags":1,"s2_open_access_pdf_url":null,"s2_open_access_landing_url":null,"s2_open_access_license":null,"s2_open_access_status":null,"pmc_open_access_pdf_url":null,"pmc_open_access_landing_url":null,"pmc_open_access_license":null,"pmc_open_access_status":null,"unpaywall_open_access_pdf_url":null,"unpaywall_open_access_landing_url":null,"unpaywall_open_access_license":null,"unpaywall_open_access_status":null,"abstract":"Design, VLSI implementation, and experimental validation of a resource-optimized machine-learning algorithm for epilepsy seizure detection is presented. The algorithm uses only signals from the frontal and the front-temporal lobes EEG electrodes while yielding a seizure detection performance competitive to the standard full EEG systems. The experimental validations prove the possibility of conducting accurate seizure detection using quickly-mountable dry-electrode headsets without the need for uncomfortable/painful through-hair electrodes or adhesive material. The compact VLSI implementation of the algorithm is also presented and resource optimization techniques are discussed. The optimized implementation is uploaded on an Actel Igloo AGL250 low-power FPGA, requires 1237 logic elements, consumes 110μW dynamic power, and yields a detection latency of 10.2μs. The measurement results from the FPGA implementation on data from 23 patients (198 seizures in total) shows a seizure detection sensitivity and specificity of 92.5% and 80.1%, respectively.","claims":[{"public_id":"cl_e21ff7ed016ed70a8d61d3cc329c0f78","status":"active","text":"Accurate seizure detection is possible with quickly mountable dry-electrode headsets without through-hair electrodes or adhesive material.","confidence":0.9,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_e21ff7ed016ed70a8d61d3cc329c0f78"},{"public_id":"cl_483dbeb34710fec4d50c08bc37941609","status":"active","text":"On data from 23 patients and 198 seizures, the FPGA implementation achieves 92.5% sensitivity and 80.1% specificity.","confidence":0.98,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_483dbeb34710fec4d50c08bc37941609"},{"public_id":"cl_421c0e1ce41a6799174a7ea47d028da0","status":"active","text":"The optimized Actel Igloo AGL250 FPGA implementation requires 1237 logic elements, consumes 110μW dynamic power, and yields a detection latency of 10.2μs.","confidence":0.98,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_421c0e1ce41a6799174a7ea47d028da0"},{"public_id":"cl_49ee89638e25254ded4f2ef5b4e24cec","status":"active","text":"Using only frontal and front-temporal EEG electrodes can achieve seizure detection performance competitive with standard full EEG systems.","confidence":0.94,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_49ee89638e25254ded4f2ef5b4e24cec"}],"concepts":[{"public_id":"co_035efe419f20cd3bbcd772297faddd9d","status":"active","name":"adhesive material","description":"Sticky material used to help secure electrodes or sensors to the body.","types":["material"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_035efe419f20cd3bbcd772297faddd9d"},{"public_id":"co_189e951832442f28b141a59fd22a7ae1","status":"active","name":"110μW dynamic power","description":"The reported dynamic power consumption of the FPGA implementation.","types":["power metric"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_189e951832442f28b141a59fd22a7ae1"},{"public_id":"co_2af1ff83d9702b16335e699a47113fb8","status":"active","name":"23 patients","description":"The number of patients whose data were used in the FPGA measurement results.","types":["dataset size"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_2af1ff83d9702b16335e699a47113fb8"},{"public_id":"co_4803c1db19b83a1c32931406b0aa3b85","status":"active","name":"sensitivity","description":"The proportion of seizure events correctly detected by the system.","types":["evaluation metric"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_4803c1db19b83a1c32931406b0aa3b85"},{"public_id":"co_4c8d0703f116bbc76e69eb31092ea6d7","status":"active","name":"198 seizures","description":"The total number of seizure events included in the evaluation data.","types":["dataset size"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_4c8d0703f116bbc76e69eb31092ea6d7"},{"public_id":"co_4e241152fc3bf22390b4a5fa6f6bed2f","status":"active","name":"seizure detection","description":"The task of identifying epileptic seizure events from recorded signals.","types":["task"],"aliases":["epilepsy seizure detection"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_4e241152fc3bf22390b4a5fa6f6bed2f"},{"public_id":"co_5a2b48a1b15065989bed0a896d104046","status":"active","name":"frontal and front-temporal lobe EEG electrodes","description":"Electroencephalography electrodes placed over the frontal and front-temporal regions.","types":["measurement setup"],"aliases":["frontal-lobe EEG","front-temporal EEG electrodes"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_5a2b48a1b15065989bed0a896d104046"},{"public_id":"co_804efa1fa1fa3f3e73da2827963e161b","status":"active","name":"1237 logic elements","description":"The reported number of logic elements required by the optimized hardware implementation.","types":["hardware resource metric"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_804efa1fa1fa3f3e73da2827963e161b"},{"public_id":"co_84a8cc0759e064cc91a25c4de883cacb","status":"active","name":"specificity","description":"The proportion of non-seizure intervals correctly classified as negative by the system.","types":["evaluation metric"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_84a8cc0759e064cc91a25c4de883cacb"},{"public_id":"co_a217be4bfc6bf4ce86ebc2c1b77058eb","status":"active","name":"dry-electrode headsets","description":"Wearable EEG headsets that use dry electrodes and do not require conductive gel.","types":["device"],"aliases":["quickly-mountable dry-electrode headsets"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_a217be4bfc6bf4ce86ebc2c1b77058eb"},{"public_id":"co_ae5efaf3127ee9002cb52f2931827569","status":"active","name":"through-hair electrodes","description":"Electrodes designed to contact the scalp through hair, often used in EEG acquisition.","types":["device component"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_ae5efaf3127ee9002cb52f2931827569"},{"public_id":"co_b82420b8fdaaaf1f61311898923c3582","status":"active","name":"10.2μs detection latency","description":"The reported time delay from input data to seizure detection output.","types":["performance metric"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_b82420b8fdaaaf1f61311898923c3582"},{"public_id":"co_c6b294755c93cd4a0418e99ba020079a","status":"active","name":"full EEG systems","description":"Conventional electroencephalography systems that use a larger set of scalp electrodes.","types":["baseline system"],"aliases":["standard full EEG systems"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_c6b294755c93cd4a0418e99ba020079a"},{"public_id":"co_df8163578d141df0db181236a0d24a1b","status":"active","name":"Actel Igloo AGL250 FPGA","description":"A low-power field-programmable gate array used for the compact hardware implementation.","types":["hardware platform"],"aliases":["Actel Igloo AGL250 low-power FPGA"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_df8163578d141df0db181236a0d24a1b"}],"external_ids":{"DOI":"10.1109/ISCAS.2019.8702211","ArXiv":null,"PubMed":null,"PubMedCentral":null,"MAG":2942968344,"DBLP":"conf/iscas/ZhanGK19","ACL":null},"open_access":{"is_open_access":false,"pdf_url":null,"landing_url":"https://sah.borca.ai/papers/155537410","source":null,"pdf_url_source":null,"license":null,"reason":"pdf_url_not_indexed"},"reference_availability":{"status":"available","references_indexed":true,"full_text_available":false,"full_text_source":null,"count_basis":"semantic_scholar_metadata","extraction_status":"not_applicable","reason":null},"source":{"provider":"episteme2","base_corpus":"semantic_scholar_dump","freshness_mode":"unknown","basis":["semantic_scholar_metadata","postgres_metadata"],"limits":["paper metadata is based on indexed upstream scholarly datasets","claims and concepts are available only for extracted papers","absence of claims or concepts means no extracted graph data is available in this response"],"status":"available","degraded":false,"degraded_reasons":[],"diagnostics":{"status":"available","degraded":false,"degraded_reasons":[],"metadata_status":"available","graph_status":"available","abstract_status":"available"},"source_flags":1},"paper_id":634566,"paper_uid":"d880ddec-f613-408b-ac52-b2895a1d35bc","canonical_identity":{"paper_id":634566,"paper_uid":"d880ddec-f613-408b-ac52-b2895a1d35bc","identity_status":"available","lookup_basis":"semantic_scholar_external_id","compatibility_path":"corpus_id"},"url":"https://sah.borca.ai/papers/155537410"}