スマートグラスを用いた心拍数モニタリングシステ
ム開発研究 ー頭部誘導心電図における新たなノイ
ズ除去アルゴリズムの提案−
著者
木原 広夢, 李 知炯
雑誌名
福岡工業大学総合研究機構研究所所報
巻
3
ページ
29-33
発行年
2020-12-25
URL
http://hdl.handle.net/11478/00001586
ࢫ࣐࣮ࢺࢢࣛࢫࢆ⏝࠸ࡓᚰᢿᩘࣔࢽࢱࣜࣥࢢࢩࢫࢸ࣒㛤Ⓨ◊✲
㸫㢌㒊ㄏᑟᚰ㟁ᅗ࠾ࡅࡿ᪂ࡓ࡞ࣀࢬ㝖ཤࣝࢦࣜࢬ࣒ࡢᥦ㸫
ᮌཎ ᗈክ㸦ᕤᏛ◊✲⛉ ሗࢩࢫࢸ࣒ᕤᏛᑓᨷ㸧
ᮤ ▱Ⅵ㸦ሗᕤᏛ㒊 ሗࢩࢫࢸ࣒ᕤᏛ⛉㸧
Development of Heart Rate Monitoring System Using Smart Glass
㸫
Proposed of the New-algorithm for Reduced Noise of Electrocardiogram Derived from Head㸫
KIHARA Hiromu 㸦Information and Systems Engineering, Graduate School of Engineering㸧 LEE Jihyoung 㸦Department of Information and Systems Engineering, Faculty of Information Engineering㸧
Abstract
We proposed the new digital filter algorithm for reduced noise of electrocardiogram㸦ECG㸧derived from the head. The proposed algorithm is based on the estimation technique of noise from one beat of ECG during exercise. The estimated noise is updated by 50% of noise derived from the latest beat and 50% of the noise derived from accumulated beat. The updated noise is devoted to ECG for detection of R-peak in next beat. In 3 healthy male participants, stationary state and 40, 50 W loads exercise using cycle ergometer for 30 seconds was performed, respectively. The results indicate that noise of ECG derived from head is decreased. In conclusion, these findings suggested that the proposed adaptive filter might be practical signal processing for reduced noise of ECG derived from head.
Keywords㸸Electrocardiogram, Adaptive-filter, Noise, Signal processing
ࡣࡌࡵ ᚰᢿᩘ㸦heart rate㸹HR㸧ࡣ㸪1 ศ㛫ᚰ⮚ࡀᢿືࡍࡿᅇ ᩘ࡛࠶ࡾ㸪1 ᢿࡢᚰ㟁ᅗ㸦electrocardiogram; ECG㸧࡛ᖜࡀ ᭱ࡶ㧗࠸R ࣆ࣮ࢡἼࢆẖᢿࡈ᳨ฟࡋ㸪ࡑࡢ㛫㛫㝸ࢆ⏝ ࠸࡚⟬ฟࡍࡿࠋHR ࡣ㸪ᚰ⮚ࡢᗣ≧ែࡢᢕᥱ㸪ࢫࢺࣞࢫࡢ ࢳ࢙ࢵࢡ㸪㐠ືᙉᗘࡢホ౯࡞ࡢࡓࡵᵝࠎ࡞ศ㔝࡛ࡼࡃ ⏝ࡉࢀ࡚࠸ࡿ(1)ࠋ㏆ᖺ࡛ࡣ㸪⮬ᕫᗣ⟶⌮ࡢࡓࡵ᪥ᖖ⏕ ά࡛⡆౽⏝࡛ࡁࡿ HR ࣔࢽࢱࣜࣥࢢࢩࢫࢸ࣒ὀ┠ࡀ 㞟ࡲࡗ࡚࠸ࡿၳ2ၴࠋ HR ࣔࢽࢱࣜࣥࢢࡢࡓࡵࡢ ECG ࡣ㸪⬚㒊ࡸᅄ⫥ࡢ⓶⾲ 㠃1 ࡘࡢ㛵㟁ᴟᚰ⮚ࢆᇶ‽ࡋࡓᕥྑ 2 ࡘࡢ㛵㟁 ᴟࢆྲྀࡾࡅ࡚ィ ࡛ࡁ㸪㛵㟁ᴟࡢ㛫ࡢ㟁ᕪࢆቑᖜࡋࡓ Ἴᙧ࡛࠶ࡿࠋ᭱㏆㸪IoT 㞟✚ᅇ㊰࡞ࡢ༙ᑟయᢏ⾡ࡢⓎᒎ క࠸㸪་⒪ᶵ㛵ࡔࡅ࡛࡞ࡃ㸪᪥ᖖ࡛ࡶHR ࢆࣔࢽࢱࣜࣥࢢ ࡛ࡁࡿ╔⾰ᆺ࡞ࡢ࢙࢘ࣛࣈࣝECG ィ ⨨ࡀ㛤Ⓨࡉࢀ ࡚࠸ࡿၳ3ၴࠋ㸪ࢫ࣮ࣜࢺࡸయࢆ㘫࠼࡚࠸ࡿே⏝ࡉࢀ ࡚࠸ࡿࡀ㸪╔⬺ࡢ↹ࢃࡋࡉࡸờࢆ࠸ࡓᚋࡢὙ℆ࡼࡿ㟁 ᴟࡢࡎࢀࡼࡿィ ⢭ᗘࡢపୗ࡞ࡢㄢ㢟ࡀṧࡗ࡚࠸ࡿࡓ ࡵ㸪⏝ࡣ㝈⏺ࡀ࠶ࡿࠋࡑࡇ࡛ᮏ◊✲࡛ࡣ㸪╔⬺ࡀᐜ࡛᫆ Ὑ℆せ࡞ࢫ࣐࣮ࢺࢢࣛࢫ࡞ࡢ࣓࢞ࢿࢆ⏝࠸ࡓ HR ࣔࢽ ࢱࣜࣥࢢࢩࢫࢸ࣒ࢆᥦࡍࡿࠋ ᥦࡋࡓࢩࢫࢸ࣒ࡣ㸪ࢫ࣐࣮ࢺࢢࣛࢫࡢࣇ࣮࣒ࣞ㟁ᴟ ࢆྲྀࡾࡅ㸪㢌㒊ࡽECG ࢆィ ࡋ㸪⟬ฟࡉࢀࡓᚰᢿᩘࢆ ࣞࣥࢬᢞᙳࡍࡿ⤌ࡳ࡛࠶ࡿࠋ୍᪉㸪㢌㒊࡛ᚓࡽࢀࡓECG ࡣ㸪2 ࡘࡢ㛵㟁ᴟࡀ㏆ࡃ㸪ᚰ⮚ࡽ㐲ࡃ㞳ࢀࡓ⨨࡛ィ ࡋ ࡚࠸ࡿࡓࡵ㸪ᚓࡽࢀࡿ㟁ᕪࡀᚤᙅ࡛࠶ࡾ㸪➽㟁࣭⬻㟁࣭య ື࡞ࡼࡿࣀࢬࡀከࡃྵࡲࢀ࡚࠸ࡿ(4-6)ࠋࡑࡢࡓࡵ㸪㢌 㒊࡛ィ ࡋࡓᚤᙅ࡞ECG ࡽ R ࣆ࣮ࢡࢆṇ☜᳨ฟࡍࡿಙ ྕฎ⌮ἲࡣ㸪ᮏࢩࢫࢸ࣒࠾࠸࡚᭱ࡶ㔜せ࡞᰾ᚰᢏ⾡ࡔ ゝ࠼ࡿࠋ ECG ࡢࣀࢬࢆ㝖ཤࡍࡿࡓࡵࡢಙྕฎ⌮᪉ἲࡋ࡚㸪࿘ Ἴᩘ≉ᚩࢆ⏝࠸ࡓࣂࣥࢻࣃࢫࣇࣝࢱ㸦band pass filter; BPF㸧 ࡸண ್ࢆ⏝࠸ࡓ࣐࢝ࣝࣥࣇࣝࢱ࡞ࡀࢃࢀ࡚࠸ࡿ(7)ࠋ ࡋࡋ㸪㐠ືࡸ᪥ᖖ⏕ά୰ࡢືࡁࡣつ๎ⓗ࡛࠶ࡿࡓࡵ ࣀࢬࡢ≉ᚩࡀண ྍ⬟࡛࠶ࡾ㸪ືసࡢࡁࡉࡼࡗ࡚ ࣀࢬࡢᖜࡀ㢌㒊ㄏᑟECG ࡢ R ࣆ࣮ࢡἼࡼࡾࡁࡃ࡞ࡿ ሙྜࡶ࠶ࡿࠋࡉࡽ㸪➽㟁ࡸ⬻㟁ࡢ࿘Ἴᩘᖏᇦࡣ㸪ᚰ㟁㔜 ࡞ࡿᖏᇦࡀ࠶ࡿ(8)ࠋࡍ࡞ࢃࡕ㸪ᐃᆺⓗ࡞ECG ಙྕฎ⌮᪉ἲ ࡔࡅ࡛㢌㒊ㄏᑟECG ࡢࣀࢬࢆ㝖ཤࡍࡿࡣ㝈⏺ࡀ࠶ࡿࠋ ࡑࡇ࡛ᮏ◊✲࡛ࡣ㸪㢌㒊ㄏᑟECG ࡽṇ☜࡞ HR ࡢ᳨ฟ ࢆ┠ᣦࡋ࡚㸪1 ᢿࡈࡢ ECG ᑐࡋ㸪ືసࡼࡗ࡚ኚືࡍ ࡿࣀࢬࡢࣃࢱ࣮ࣥࢆᢳฟཬࡧᏛ⩦ࡋ࡚ࣀࢬࢆῶࡽࡍ᪂ ࡓ࡞ࣇࣝࢱࣝࢦࣜࢬ࣒ࢆᥦࡋ㸪ࣉࣟࢢ࣒ࣛࡋࡓࠋࡲ ࡓ㸪㐠ືㄢ㢟୰ィ ࡋࡓ㢌㒊ㄏᑟECG ࢆ⏝࠸࡚㸪ᥦࡋ ࡓࣝࢦࣜࢬ࣒ࡢホ౯ࢆ⾜ࡗࡓࠋ
ᮌཎ ᗈክ㸪ᮤ ▱Ⅵ
2. ィ ⨨㐺ᛂࣇࣝࢱ
2.1 㢌㒊ㄏᑟ ECG ィ ⨨
ᚤᙅ࡞ಙྕࢆィ ࡍࡿࡓࡵ㸪ィ ⨨ࡣධຊࣥࣆ࣮ࢲ ࣥࢫࡢ㧗࠸ィࣥࣉ㸦AȍINA116, Texas Instruments, USA㸧ࢆ⏝࠸࡚ヨసࡋࡓࠋࡲࡓ㸪ECG ࡢ࿘Ἴᩘᖏᇦ࡛࠶ࡿ 0.01 ~ 250 Hz ࡢಙྕࢆ 50,000 ಸቑᖜࡋ࡚࠸ࡿࠋ 2.2 ᪂ࡓ࡞ࣇࣝࢱࣝࢦࣜࢬ࣒ ᥦࡍࡿ᪂ࡓ࡞ࣇࣝࢱࣝࢦࣜࢬ࣒㸦௨ୗ㸪㐺ᛂࣇࣝ ࢱ㸧ࡣ㸪1 ᢿࡈࡢࣀࢬࢆᏛ⩦ࡍࡿタィ࡞ࡗ࡚࠾ࡾ㸪ᅗ 1 ࡢࡼ࠺࡞ᡭ㡰࡛ࣇࣝࢱฎ⌮ࢆ⾜ࡗ࡚࠸ࡿࠋ ࣝࢦࣜࢬ࣒ࡢጞࡲࡾ࡛ࡣ BPF ฎ⌮ࡉࢀࡓࢹ࣮ࢱᑐࡋ ࡚㸪R ࣆ࣮ࢡࡔண ࡉࢀࡿⅬࢆᇶ‽ࡋ࡚ 1 ᢿࡈ ษࡾศࡅ㸪Ᏻ㟼௬ᐃࡉࢀࡿึᮇ㛫ࡽ 5 ᢿࡢᖹᆒ್ ࢆECG ࡢᩍᖌࢹ࣮ࢱ㸪ཬࡧ೫ᕪ್ࢆࣀࢬࡢᩍᖌࢹ࣮ࢱ ࡋ࡚ᢳฟࡍࡿࠋḟࡢẁ㝵࡛ࡣ㸪ࣀࢬࢆᏛ⩦ࡋ࡚㝖ཤࡍࡿࠋ ࣀࢬࡢᏛ⩦㝖ཤ᪉ἲࡘ࠸࡚ࡣ㸪ᅗ 1 ࡢ㉥࠸Ⅼ⥺࡛ ⾲ࡋ࡚࠸ࡿࠋࣀࢬࡢᏛ⩦ࢆ⾜࠺ࡓࡵ 1 ᢿ┠࠾ࡅࡿᏛ ⩦ࣀࢬ㸦f㸦k㸧㸧ࢆ⟬ฟࡋࡓࠋ⟬ฟࡣ㸪➽㟁ࡸయື࣮ ࢳࣇࢡࢺ࡞ࡢ㐠ືࡼࡗ࡚⏕ࡌࡿ㐠ືࣀࢬ㸦g㸦k㸧㸧㸪 Ᏻ㟼ࡢ㢌㒊ㄏᑟECG ࡢ 1 ᢿࡢᇶ‽࡞ࡿᖹᆒ㸦tECG㸧ཬ ࡧ➽㟁௨እࡢ㟁☢Ἴࡸ⬻Ἴ࡞ࡢࣛࣥࢲ࣒ࣀࢬࢆ⟬ฟࡍ ࡿࡓࡵࡢᶆ‽೫ᕪ㸦nECG㸧ࢆ⏝࠸ࡓࠋ㐠ືࣀࢬ㸦g㸦k㸧㸧 ࡣ㐠ືࡢ㢌㒊ㄏᑟECG ࡽᩍᖌࢹ࣮ࢱࡢᕪศࢆ⾜࠺ࡇ ࡛㸪Ᏻ㟼ࡣ⏕ࡌ࡞࠸㐠ືࡢࡳ⏕ࡌࡿࣀࢬࢆ⟬ฟ ࡋࡓࠋᕪศࢆ⾜࠺࠶ࡓࡾ㸪㐠ືᏳ㟼ࡢ㢌㒊ㄏᑟECG ࡢ1 ᢿࡢ㛫㝸ࡣ␗࡞ࡿࡓࡵ㸪1 ᢿࡈࡢ㛫㝸ࢆ୍ᐃࡍࡿ⥺ ᙧ⿵㛫ࢆ⾜ࡗࡓࠋ㸯ᢿ┠ࡢ࣮࣋ࢫࣛࣥ࡞ࡿf㸦k㸧ࡣ g㸦k㸧 ᑐࡋ㸪Ᏻ㟼ࡸ㐠ືࡶඹ㏻ࡋ࡚⏕ࡌࡿࣛࣥࢲ࣒ࣀ ࢬ㸦nECG㸧ࢆ 1 ᑐ 1 ࡢ㔜ࡳ࡛ྜᡂࡋ⟬ฟࡋࡓࠋ2 ᢿ┠௨㝆 ࡢᏛ⩦ࣀࢬ㸦f㸦k㸧㸧ࡣ㸪1 ᢿ๓ࡢᏛ⩦ࣀࢬ㸦f㸦k - 1㸧㸧 g㸦k㸧ࢆຍ⟬ࡍࡿࡇ࡛ 1 ᢿ๓ࡲ࡛ࡢࣀࢬࡢሗࡀ ✚ࡉࢀࡓᏛ⩦ࣀࢬࢆ⟬ฟࡋ࡚࠸ࡿ㸬 ᭱ᚋࣀࢬࡢ㝖ཤ᪉ἲࡘ࠸࡚࡛࠶ࡿࠋ㐠ືࡢ㢌㒊 ㄏᑟECG ࡢ 1 ᢿᑐࡋ㸪ࣀࢬࡢᏛ⩦ࡼࡾᚓࡽࢀࡓ f㸦k㸧 ࢆᕪศࡍࡿࡇ࡛㸪㐠ືࡢ㢌㒊ㄏᑟECG ࡽࣀࢬࡀ㝖 ཤࡉࢀࡓ㢌㒊ㄏᑟECG㸦y㸦k㸧㸧ࢆ⟬ฟࡋ࡚࠸ࡿࠋࡲࡓ㸪Ꮫ ⩦ࣀࢬ㸦f㸦k㸧㸧ࡣᕪศࡢ๓㸪ධຊࢹ࣮ࢱ࠾ࡅࡿ㐠ື ࡢ㢌㒊ㄏᑟECG ྠࡌ㛫㍈ࡍࡿࡓࡵ⥺ᙧ⿵㛫ࢆ⾜ ࡗ࡚࠸ࡿࠋ 3. ᐇ㦂 3.1 ᛶ⬟ホ౯ヨ㦂 㢌㒊ㄏᑟECG ィ ⨨ࢆ⏝࠸࡚㢌㒊ㄏᑟ ECG ࡢィ ཬ ࡧ㸪R ࣆ࣮ࢡࡢ᳨ฟ⋡ࡢ⟬ฟࡢࡓࡵ㸪ECG ࡽᚓࡽࢀࡿ R ࣆ ࣮ ࢡ ᙉ ࠸ ┦ 㛵 㛵 ಀ ࢆ ᣢ ࡘ ග 㟁 ᐜ ✚ ⬦ Ἴ 㸦 photo-plethysmogram; PPG㸧ࡢྠィ ࢆ⾜ࡗࡓࠋࡲࡓ㸪᪥ᖖ⏕ ά୰ࡢᵝࠎ࡞యືࢆ⪃៖ࡋ㸪Ᏻ㟼≧ែ⮬㌿㌴࢚ࣝࢦ࣓࣮ ࢱࢆ⏝࠸ࡓ㐠ື㈇Ⲵㄢ㢟ࢆᐇࡋࡓࠋᮏᐇ㦂ࡣ㸪࣊ࣝࢩࣥ࢟ ᐉゝࡢ⢭⚄๎ࡾ㸪ᑐ㇟⪅ࡣᮏ◊✲㛵ࡍࡿ༑ศ࡞ᐇ㦂 ᪨ㄝ᫂ࢆ⾜࠸㸪ཧຍࡢ௵ពᛶࢆᩥ᭩࠾ࡼࡧཱྀ㢌࡚ㄝ ᫂ࡋ㸪᭩㠃࡚ྠពࢆᚓࡓୖ࡛ᐇࡋࡓࠋ 3.2 ィ ᑐ㇟㔞 㢌㒊ㄏᑟECG ࡣࣇ࢛࣮࣒ࢸ࣮ࣉࢱࣉ Ag࣭AgCl ࡛࠶ࡿ 3 ࡘࡢࢣࣥࢻ࣮ࣝ㟁ᴟࣝ࣎ࢆࢫ࣐࣮ࢺࢢࣛࢫ㢌㒊ࡢ᥋ ゐ⨨ࢆ⪃៖ࡋ㸪୧⪥ྲྀࡾࡅ㸪㢌㒊ㄏᑟECG ィ ⨨ࢆ⏝ࡋ㸪ィ ࢆ⾜ࡗࡓࠋࡲࡓ㸪PPG ࡣᣦᑤ㒊ᑕᆺࡢ ⥳ගࢭࣥࢧࣔࢪ࣮ࣗࣝ㸦525 nm㸧ࢆྲྀࡾࡅ࡚ィ ࢆ⾜ࡗ ᅗ1 ᥦࡋࡓࣝࢦࣜࢬ࣒ࡢࣇ࣮ࣟࢳ࣮ࣕࢺ Fig. 1 Flow chart of proposed algorithm㸬
ᅗ2 㟁ᴟཬࡧࢭࣥࢧࡢ⨨㸦a㸧 㐠ື㈇Ⲵㄢ㢟㸦b㸧 Fig. 2 The position of electrode and sensor㸦left; a㸧,
ࡓࠋECG ࡢ㟁ᴟཬࡧ PPG ࡢ⥳ගࢭࣥࢧࣔࢪ࣮ࣗࣝࡣఙ⦰ ᛶࢸ࣮ࣉࢆ⏝ࡋ㸪ᅛᐃࡋࡓ㸦ᅗ2㸦a㸧ཧ↷㸧ࠋྛィ ⨨ ࡽࡢࢼࣟࢢಙྕࡣ1 kHz ࡛ࢧࣥࣉࣜࣥࢢࢆ⾜ࡗࡓࠋ 3.3 ᐇ㦂ᡭ㡰 ィ ᐇ㦂ࡣᗣᡂே⏨ᛶ 3 ྡࢆ⿕㦂⪅ࡋ㸪ᐊ ࡀ⣙ 24.3Υಖࡓࢀࡓ⚟ᒸᕤᴗᏛሗࢩࢫࢸ࣒ᕤᏛ⛉ B7280 ࡢ◊✲ᐊෆⶶࡉࢀࡓࢩ࣮ࣝࢻ࣮࣒ࣝ㸦㐽᩿40 dB㸧ࢆ⏝ ࡋࡓࠋ⿕㦂⪅ࡣࢩ࣮ࣝࢻ࣮࣒ࣝධᐊᚋ㟁ᴟཬࡧࢭࣥࢧࢆ ╔ࡋ࡚ᚅᶵࡋࡓࠋࡑࡢᚋ⮬㌿㌴࢚ࣝࢦ࣓࣮ࢱ㌴ࡋ㸪Ᏻ 㟼ཬࡧ㐠ື㈇Ⲵࡢᐇ㦂ㄢ㢟ࡶ㢌㒊ㄏᑟECG ཬࡧ PPG ࡢྠィ ࢆ⾜ࡗࡓࠋᐇ㦂ㄢ㢟ࡣ㸪Ᏻ㟼㸦0 W㸧㸪40 W㸪50 W㸪40 W ࡢ㐠ື㈇Ⲵ㸪Ᏻ㟼㸦0 W㸧ࡢ㡰ྛ 30 ⛊ࡢィ ࢆ ᐇࡋࡓ㸦ᅗ2㸦b㸧ཧ↷㸧ࠋ㐠ື㈇Ⲵᚋࡣ⮬㌿㌴࢚ࣝࢦ࣓࣮ ࢱࡽ㝆㌴ࡋ㸪㟁ᴟཬࡧࢭࣥࢧࢆྲྀࡾእࡋ࡚ᐇ㦂ࢆ⤊ࡋ ࡓࠋ㐠ື㈇Ⲵᚋࡣ㐠ື㈇Ⲵᑐࡍࡿ⿕㦂⪅ࡢほⓗ࡞ឤ ぬࢆ☜ㄆᚋ㸪⿕㦂⪅ࡢពᚿᚑࡗ࡚ᐇ㦂ࡢ⥅⥆ཬࡧ୰Ṇࢆ ุ᩿ࡋࡓࠋ㐠ື㈇Ⲵࡢ㌟యࡢືࡁ㛵ࡋ࡚ࢥࣥࢺ࣮ࣟࣝ ࡣ࡞ࡗࡓࠋࡲࡓ㸪㐠ື㈇Ⲵㄢ㢟ࡣ⣙2 ศ㛫࡛࠶ࡗࡓࠋ 3.4 ࢹ࣮ࢱゎᯒ ィ ࡋࡓ㢌㒊ㄏᑟECG ࡣ 10 ಸࡢቑᖜ㸪8 ~25 Hz ࡢ BPF ࡢཬࡧ㐺ᛂࣇࣝࢱࡢಙྕฎ⌮ࢆ⾜ࡗࡓࠋィ ࡋࡓPPG ࡣ0.3 ~ 30 Hz ࡢ BPF ࡢಙྕฎ⌮ࢆ⾜ࡗࡓࠋࡲࡓ㸪㜈್ࢆ ⏝ࡋࡓR ࣆ࣮ࢡࡢ᳨ฟࣝࢦࣜࢬ࣒ࢆ⏝ࡋ㸪R ࣆ࣮ࢡࡢ ᳨ฟࢆ⾜ࡗࡓࠋR ࣆ࣮ࢡࡢ᳨ฟ⢭ᗘࡢᣦᶆ࡞ࡿ᳨ฟ⋡ 㸦detection ratio; DR㸧ࡣ㸪ṇ☜᳨ฟࡉࢀࡓ R ࣆ࣮ࢡ㸦true
positive; TP㸧㸪ㄗ᳨ฟ㸦false positive; FP㸧㸪ᮍ᳨ฟ㸦false negative; FN㸧㸪PPG ࡽᚓࡽࢀࡓṇ☜࡞ R ࣆ࣮ࢡࡢᩘࢆ⏝ࡋ㸪௨ ୗࡢィ⟬ᘧࢆ⏝࠸࡚⟬ฟࡉࢀ࡚࠸ࡿ(9-10)ࠋ DR = ቆ1 െ ቀ ிାிே ்௧ ௨ ௦ቁቇ × 100 㸦1㸧 4. ⤖ᯝ 4.1 㐺ᛂࣇࣝࢱࡢ㐺⏝๓ᚋ࠾ࡅࡿ㢌㒊ㄏᑟ ECG ࡢẚ㍑ ᅗ3 ࡣ 8 ~ 25 Hz ࡢ BPF 㐺ᛂࣇࣝࢱࡢಙྕฎ⌮ࢆ⾜ ࠸㸪㢌㒊ㄏᑟECG Ἴᙧࢆ㐠ື㈇Ⲵࡈẚ㍑ࡋࡓ⤖ᯝ࡛࠶ ࡿࠋ0 W㸪40 W㸪50 W㸪40W㸪0 W ࡢ࠸ࡎࢀࡢ㐠ື㈇Ⲵ࠾ ࠸࡚ࡶ㢌㒊ㄏᑟECG Ἴᙧࡢ୰ኸᏑᅾࡍࡿࣀࢬࡢࡁࡉ ࡣ㐺ᛂࣇࣝࢱࡢ㐺⏝๓ẚ㸪ᑠࡉࡃ࡞ࡗࡓࠋ 4.2 㐺ᛂࣇࣝࢱ BPF ࠾ࡅࡿ R ࣆ࣮ࢡࡢ᳨ฟ⋡ ⾲1 ࡣ㐺ᛂࣇࣝࢱ BPF ࠾ࡅࡿ R ࣆ࣮ࢡࡢ᳨ฟ⋡ࢆ ࡲࡵࡓ⤖ᯝ࡛࠶ࡿࠋ㐺ᛂࣇࣝࢱBPF ࢆ㐺⏝ࡋࡓ㢌㒊 ㄏᑟECG ࡢ R ࣆ࣮ࢡࡢ᳨ฟ⋡ኚࡣぢࡽࢀ࡞ࡗࡓࠋ 5. ⪃ᐹ ᮏ◊✲ࡢ┠ⓗࡣ㸪᪥ᖖ⏕ά୰ࡢ⡆౽࡞HR ࣔࢽࢱࣜࣥࢢࡢ ࡓࡵࡢࢫ࣐࣮ࢺࢢࣛࢫࢆ⏝࠸ࡓ HR ࣔࢽࢱࣜࣥࢢࢩࢫࢸ࣒ ࡢ㛤Ⓨࢆ┠ᣦࡋ㸪᪥ᖖⓗ࡞ண ྍ⬟࡞ࣀࢬࡶᑐᛂྍ ⬟࡞ィ ⎔ቃࡼࡗ࡚␗࡞ࡿࣀࢬࢆᏛ⩦ࡍࡿ㐺ᛂࣇࣝ ࢱࢆ㛤Ⓨࡋ㸪㐺ᛂࣇࣝࢱࡢ㐺ᛂ๓ᚋ࠾ࡅࡿ㢌㒊ㄏᑟ ECG ࡢẚ㍑ࢆ⾜࠺ࡇ࡛ࣀࢬ㝖ཤࡢᛶ⬟ࡘ࠸᳨࡚ウࢆ ᅗ3 ࣂࣥࢻࣃࢫࣇࣝࢱཬࡧ㐺ᛂࣇࣝࢱࢆ㐺⏝ࡋࡓᏳ㟼㐠ື㈇Ⲵ㸦0 W㸦a㸧㸪40 W㸦b㸧㸪50 W㸦c㸧㸪40 W㸦d㸧㸪 0 W㸦e㸧㸧ࡢ㢌㒊ㄏᑟᚰ㟁ᅗἼᙧ
Fig. 3 The Waves of ECG derived from the head adapted band pass filter and adaptive filter during stationary state and exercise load 㸦0 W㸦a㸧㸪40 W㸦b㸧㸪50 W㸦c㸧㸪40 W㸦d㸧㸪0 W㸦e㸧㸧.
ᮌཎ ᗈክ㸪ᮤ ▱Ⅵ ⾜࠺࡛࠶ࡗࡓࠋᅗ3 ࡢ⤖ᯝࡼࡾ㸪࠸ࡎࢀࡢ㐠ື㈇Ⲵ࠾ ࠸࡚ࡶ㐺ᛂࣇࣝࢱࢆ㐺⏝ࡋࡓ㢌㒊ㄏᑟECG ࡢἼᙧࡀࣀ ࢬࢆపῶࡋ࡚࠸ࡿࢆ☜ㄆ࡛ࡁࡓࠋࡇࡢࡇࡼࡾ㸪ᅇᵓ ⠏ࢆ⾜ࡗࡓ㐺ᛂࣇࣝࢱࡣ㸪⬚㒊ࡽィ ࡉࢀࡓECG ⏝࠸ࡽࢀࡿ࿘Ἴᩘ≉ᛶࢆ⏝ࡋࡓBPF ࡼࡾ㸪ࣀࢬࢆపῶ ࡛ࡁࡿࡀ☜ㄆࡉࢀࡓࠋࡉࡽ㸪㐠ື㈇Ⲵࢆኚືࡉࡏࡓ㐃⥆ ⓗ࡞Ἴᙧࡘ࠸࡚ࡶࣀࢬࢆపῶ࡛ࡁࡿࡇࡀ☜ㄆࡉࢀ ࡓࠋࡇࢀࡣ㸪ᵓ⠏ࡋࡓ㐺ᛂࣇࣝࢱࡀ୍ᢿ๓ࡲ࡛ࡢࣀࢬࡢ ሗཬࡧ࣮࣋ࢫࣛࣥࡢ㢌㒊ㄏᑟECG ࢆ 1 ᑐ 1 ࡢྜ࡛ྜ ᡂࡍࡿฎ⌮ࢆ⾜ࡗ࡚࠸ࡿࡓࡵ㸪ィ ⎔ቃኚࡀ⏕ࡌ࡚ࡶ ᵝࠎ࡞ࣀࢬࢆ㝖ཤ࡛ࡁ࡚࠸ࡿ⪃࠼ࡽࢀ㸪ࣀࢬࡢᏛ⩦ ຠᯝࡀ࠶ࡿࢆ♧ࡋࡓࠋࡋࡓࡀࡗ࡚㸪ᵓ⠏ࡋࡓ㐺ᛂࣇࣝࢱ ࡣࡇࢀࡲ࡛⏝࠸ࡽࢀ࡚ࡁࡓBPF ௨ୖࡢࣀࢬࢆపῶ࡛ࡁࡿ ࡇࡀ᫂ࡽ࡞ࡾ㸪㢌㒊ㄏᑟECG ࠾ࡅࡿࣀࢬ㝖ཤࡢ ࡓࡵࡢࢹࢪࢱࣝಙྕฎ⌮᪉ἲࡢ㸯ࡘࡋ࡚᭷⏝࡛࠶ࡿ⪃ ࠼ࡽࢀࡿࠋ ⾲1 ࡢ⤖ᯝࡽ㸪BPF 㐺ᛂࣇࣝࢱࢆࡑࢀࡒࢀ㐺⏝ࡋ ࡓ㢌㒊ㄏᑟECG ࡢ R ࣆ࣮ࢡࡢ᳨ฟ⋡ኚࡀ࡞ࡗࡓࡀ ☜ㄆࡉࢀࡓࠋᅇ⏝ࡋࡓR ࣆ࣮ࢡࡢ᳨ฟ᪉ἲࡀ୍ᐃࡢ㜈 ್ࢆ⏝࠸࡚࠸ࡿࡇࡀせᅉ࡛࠶ࡿ⪃࠼ࡽࢀࡿࠋ㐠ື㈇Ⲵ ࡼࡗ࡚R ࣆ࣮ࢡࡢ㧗ࡉࡸࣀࢬࡢࡁࡉࡢኚື୍ᐃࡢ 㜈್࡛ࡣᑐᛂ࡛ࡁࡎ㸪㜈್ࡢ㧗ࡉࢆ㉸࠼࡞ࡗࡓᮍ᳨ฟࡢR ࣆ࣮ࢡࡀࣀࢬࡋ࡚㝖ཤࡉࢀ㸪㜈್ࡢ㧗ࡉࢆ㉸࠼ࡓ୍㒊 ࡢࣀࢬࡣṇࡋ࠸R ࣆ࣮ࢡࡋ࡚ㄗ᳨ฟ࡞ࡾ㸪ࣀࢬ ࡋ࡚ࡢ㝖ཤࡀࡉࢀ࡞ࡗࡓ⪃࠼ࡽࢀࡿࠋࡲࡓ㸪ᅗ㸱ࡢ 10~15 ⛊༊㛫࣭130~135 ⛊༊㛫ࡢࡼ࠺ R ࣆ࣮ࢡࡀᑠࡉࡃィ ࡉࢀࡓሙྜࡣ㸪ᥦࡋࡓ㐺ᛂࣇࣝࢱࡀࣀࢬࡋ࡚ฎ ⌮ࡍࡿࡢ࡛㸪R ࣆ࣮ࢡࡢ㧗ࡉࡀࡶࡗᑠࡉࡃ࡞ࡾ㸪ྠࡌ⨨ ࡛ࡢ᳨ฟ࣑ࢫࡀⓎ⏕ࡍࡿࠋᚑࡗ࡚㸪㧗࠸⢭ᗘࡢ㹐ࣆ࣮ࢡࢆ᳨ ฟࡓࡵ㸪R ࣆ࣮ࢡࡀฟࡿ㛫ࡢ๓ᚋࡢἼᙧ≉ᚩࢆศᯒࡋ࡚ R ࣆ࣮ࢡࢆ᥎ᐃࡍࡿ᪂ࡓ࡞᳨ฟ᪉ἲࡀᚲせ⪃࠼ࡽࢀࡿࠋ ᚋࡢㄢ㢟ࡋ࡚ 2 ࡘࡢࡀᣲࡆࡽࢀࡿࠋ㸯ࡘ┠ࡣࣀ ࢬࡢᏛ⩦ຠᯝࡘ࠸࡚࡛࠶ࡿࠋᅇ㸪1 ᢿ๓ࡲ࡛ࡢࣀࢬཬ ࡧ࣮࣋ࢫࣛࣥࡢ㢌㒊ㄏᑟECG ࢆ 1 ᑐ 1 ࡢྜ࡛ྜᡂࡍࡿ ฎ⌮ࢆ⾜ࡗ࡚࠸ࡿࡇࡼࡾ㸪ࣀࢬࡢᏛ⩦ຠᯝࢆ♧ࡋࡓ ࡀ㸪Ꮫ⩦ࡢ㔜ࡳࡀ1 ᑐ 1 ࡛ࡢ᳨ドࡋ⾜ࡗ࡚࠸࡞࠸ࡓࡵ㸪 ࣀࢬ㝖ཤᑐࡋ㸪᭦࡞ࡿ㧗ᛶ⬟ࡢ㐺ᛂࣇࣝࢱᵓ⠏ྥ ࡅ㸪」ᩘࡢ㔜ࡳ࡛ࡢẚ㍑᳨ウࢆ⾜࠺ᚲせࡀ࠶ࡿ⪃࠼ࡽࢀ ࡿࠋ2 ࡘ┠ࡣ R ࣆ࣮ࢡࡢ᳨ฟ⋡ྥୖྥࡅࡓ᳨ฟ᪉ἲࡢᨵ ၿࡘ࠸࡚࡛࠶ࡿࠋ㐠ື㈇Ⲵࡀࡁࡃ࡞ࡿࡘࢀR ࣆ࣮ࢡ ࡢ㧗ࡉཬࡧࣀࢬࡢࡁࡉࡶቑ࠼ࡿࠋࡑࡢࡓࡵ㸪⌧ᅾ⏝ࡋ ࡚࠸ࡿ୍ᐃࡢ㜈್ࢆ⏝ࡋࡓR ࣆ࣮ࢡࡢ᳨ฟ࡛࠶ࡿ㸪⎔ ቃࡢኚᑐࡋ㸪R ࣆ࣮ࢡࡢ᳨ฟࡀᑐᛂ࡛ࡁ࡞࠸ࠋࡋࡓࡀࡗ ࡚㸪ィ ⎔ቃ㐺ࡋࡓ㜈್ኚ᭦ྍ⬟࡞Ꮫ⩦ຠᯝࡢ࠶ࡿ R ࣆ࣮ࢡࡢ᳨ฟ᪉ἲࡘ࠸᳨࡚ウࢆ⾜࠺ᚲせࡀ࠶ࡿ⪃࠼ࡽ ࢀࡿࠋ 6. ⤖ゝ ᮏ◊✲࡛ࡣ㸪㢌㒊ㄏᑟECG ࡢィ ⎔ቃࡼࡗ࡚␗࡞ࡿࣀ ࢬࢆᏛ⩦ࡍࡿ㐺ᛂࣇࣝࢱࢆ㛤Ⓨࡋ㸪㐺ᛂࣇࣝࢱࡢ㐺 ⏝๓ᚋ࠾ࡅࡿ㢌㒊ㄏᑟECG ࡢẚ㍑ࢆ⾜࠸㸪ࣀࢬ㝖ཤᛶ ⬟ࡘ࠸᳨࡚ウࢆ⾜ࡗࡓࠋࡑࡢ⤖ᯝ㸪ᵓ⠏ࡋࡓ㐺ᛂࣇࣝࢱ ࡣࡇࢀࡲ࡛ECG ࡢࣀࢬ㝖ཤ⏝ࡉࢀ࡚ࡁࡓ࿘Ἴᩘ≉ᛶ ࢆ⏝࠸ࡓࢹࢪࢱࣝಙྕฎ⌮௨ୖࡢࣀࢬ㝖ཤᛶ⬟ࢆ♧ࡋ㸪 㢌㒊ㄏᑟECG ࠾ࡅࡿ᪂ࡓ࡞ࣀࢬ㝖ཤࣇࣝࢱࡋ࡚᭷ ⏝࡛࠶ࡿࡇࡀ♧၀ࡉࢀࡓࠋ୍᪉࡛㸪ᵓ⠏ࡋࡓ㐺ᛂࣇࣝࢱ ࡼࡗ࡚㸪R ࣆ࣮ࢡࡢ᳨ฟ⋡ྥୖࡣぢࡽࢀ࡞ࡗࡓࡀ㸪ࣀ ࢬ㝖ཤࡼࡾ㸪ࡇࢀࡲ࡛☜ㄆࡀᅔ㞴࡛࠶ࡗࡓR ࣆ࣮ࢡࢆከ ࡃ☜ㄆࡍࡿࡇࡀ࡛ࡁࡓࠋࡑࡢࡓࡵ㸪ᮏ◊✲ࡢ᳨ドࡣ㢌㒊ㄏ ᑟECG ࡢ R ࣆ࣮ࢡࡢ᳨ฟ⋡ࢆྥୖࡉࡏࡿ୍Ṍ࡞ࡾ㸪᪥ᖖ ⏕ά୰ࡢ HR ࣔࢽࢱࣜࣥࢢࢩࢫࢸ࣒ࢆᐇ⌧ࡍࡿ┠ᶆ㏆࡙ ࠸ࡓ⪃࠼ࡽࢀࡿࠋ ㅰ㎡ ᮏ◊✲ࡣ㸪ᮏᏛሗ⛉Ꮫ◊✲ᡤࡢᖹᡂ31 ᖺᗘ◊✲㈝㸦◊ ✲ࣥࢭࣥࢸࣈไᗘ㸧ཬࡧ࢝ࢩ࢜⛉Ꮫ⯆㈈㛸ࡢ◊✲ຓ ᡂࡼࡾᐇࡋࡓࡶࡢ࡛࠶ࡿࠋࡇࡇㅰពࢆ⾲ࡍࠋࡉࡽ㸪 ᐇ㦂ࡈ༠ຊ㡬࠸ࡓฟཱྀಟᖹ㸦⚟ᒸᕤᴗᏛᕤᏛ◊✲⛉ಟ ኈㄢ⛬ 1 ᖺ⏕㸧ྩཬࡧᮏ◊✲ཧຍࡋࡓᏛ⏕ㅖẶឤㅰࢆ ⾲ࡍࠋ ᩥ ⊩ (1) ᑠ㔝ᑎ Ꮥ୍࣭ᐑୗ ṇ㸸ࠕ㌟ᣢஂᛶ㐠ື࠾ࡅࡿほⓗᙉᗘᐈ ほⓗᙉᗘࡢᑐᛂᛶ : Rating of perceived exertion ࡢほⅬࡽࠖ, య⫱ Ꮫ◊✲, Vol. 21, No. 4, pp.191-203 (1976)
(2) S. Muangsrinoon and P. Boonbrahm, “Burn in Zone: Real time HeartRate monitoring for physical activity,” International Joint Conferenceon Computer Science and Software Engineering, pp.1-6 (2017)
(3) Jerald Yoo, L. Yan, Seulki Lee, Hyejung Kim, and Hoi-Jun Yoo, “A Wearable ECG Acquisition System With Compact Planar-Fashionable Circuit Board-Based Shirt”, IEEE Transactions on Information Technology in Biomedicine 2009, Vol.13, No.6, pp.897-902 (2009)
(4) Andres L. Bleda, Rafael Maestre, Björn Schmitz, Christian Hofmann, Jose M. Nacenta, Guadalupe Santa, Soledad Pellicer, and Vivien Melcher, “Electrical cardiac monitoring in the head for helmet applications”, 2015 Computing in Cardiology Conference, pp.413-416 (2015)
(5) David Da He, E. S. Winokur, and C. G. Sodini, “A continuous, wearable, and wireless heart monitor using head ballistocardiogram (BCG) and head electrocardiogram (ECG)”, Proceeding IEEE Engineering in Medicine and Biology conference 2011, pp.4729-4732 (2001)
(6) Z. Sijerci c and G. Agarwal, “Tree structured filter bank for time-frequency decomposition of EEG signals,” Proceedings of 17th International Conference of the Engineering in Medicine and Biology Society, vol.2,
⾲1 㐺ᛂࣇࣝࢱ BPF 㐺⏝ࡋࡓ㢌㒊ㄏᑟᚰ㟁ᅗ࠾ࡅ ࡿR ࣆ࣮ࢡࡢ᳨ฟ⋡ࡢ⤖ᯝ
Table 1. Detection ratio of R-peak from ECG derived from head adapted adaptive filter and BPF
Total number of R-peak False Positive False Negative Detection ratio [%] Band pass filter 624 75 16 85.42 Adaptive filter 624 75 16 85.42
pp.991–992 (1995)
(7) N.V. Thakor, Y.-S. Zhu, “Applications of adaptive filtering to ECG analysis: noise cancellation and arrhythmia detection”, IEEE Transactions on Biomedical Engineering, Vol.38, pp.785 - 794 (1991)
(8) XIAOJUN. Z, XIULI. M. & YANG, Li, “An adaptive threshold algorithm based on wavelet in QRS detection”, 2014 International Conference on Audio, pp.858-862 (2014)
(9) Y. Wang, C. J. Deepu, and Y. Lian, “A computationally efficient QRS detection algorithm for wearable ECG sensors” 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, pp. 5641-5644 (2011)
(10) F. Zhang and Y. Lian, “QRS detection based on multiscale mathematical morphology for wearable ECG devices in body area networks”, IEEE Transactions on Biomedical Circuits and Systems, Vol.3, No.4, pp.220-228 (2009)
