JP2004212120A

JP2004212120A - Measuring instrument having measured value estimating function and measuring method using the same

Info

Publication number: JP2004212120A
Application number: JP2002379823A
Authority: JP
Inventors: Masashige Fusamoto; 正滋房本; Masakazu Tsuchiya; 正和土谷; Haruki Oishi; 晴樹大石
Original assignee: Wako Pure Chemical Industries Ltd
Current assignee: Fujifilm Wako Pure Chemical Corp
Priority date: 2002-12-27
Filing date: 2002-12-27
Publication date: 2004-07-29

Abstract

<P>PROBLEM TO BE SOLVED: To provide a measuring instrument for estimating a time arriving at a definite threshold value without extending a measuring time using the data of a change in transmitted light or absorbancy obtained within the measuring time with respect to a specimen not arriving at the definite threshold value within the measuring time. <P>SOLUTION: A ratio of quantity of transmitted light or absorbancy after a sample to be inspected is mixed with a gelling or coloring matter generating reagent but before changed by reaction with quantity of transmitted light or absorbancy changing with the elapse of time or the change of the ratio with the elapse of time (time course) caused by the reaction of the sample to be inspected is monitored to measure the time required until the ratio arrives at a predetermined definite value (gelling or coloring matter generation judge threshold value). Alternately, in such a case that the ratio does not arrives at the judge threshold value at the point of time when the measuring time is completed, the monitored time couse is extended on and after the measuring time by estimate calculation to calculate an estimate time course curve and the time required until the estimate time course curve arrives at the judge threshold value is calculated to measure the specimen not arriving at the definite threshold value within the measuring time for a short time. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

【０００１】
【発明の属する技術分野】
この発明は、測定対象物質濃度の測定に使用される測定値推定機能を持つ生化学的若しくは免疫学的測定装置及び測定方法に係り、詳記すれば、測定時間内に一定の閾値に到達しないため測定に長時間要する検体について、迅速に測定できるようにした測定値推定機能を持つ生化学的若しくは免疫学的測定装置及び測定方法に関する。
【０００２】
【従来技術】
本発明は、主としてエンドトキシン、β―グルカン又はペプチドグリカンの測定に使用されるが、この代表例であるエンドトキシンは、発熱性物質の代表的なものである。このエンドトキシンが混入した血液、注射液などが生体内に注入されると強い発熱やショックなどの重篤な副作用を引き起こすことが知られている。
【０００３】
近年、エンドトキシンの測定法としては、カブトガニ血球成分、例えばリムルス・アメーボサイト・ライセート（以下ＬＡＬと略す。）がエンドトキシンと反応してゲル化することが見出され、これを応用したエンドトキシンの測定法が開発されている。
【０００４】
しかしながら、この従来のエンドトキシンの測定法は、正確で安定した測定値を得るという点に問題があったので、本出願人は、各々にゲル化試薬を混合した２以上の被検試料に各々光線を照射し、混合後の透過光量の最大値（反応により変化する前の値）と経時的に変化する透過光量との比率を各々測定し、混合時点から該比率が急激な減少を開始し始めた後の値であってかつ予め設定された一定値(ゲル化判定閾値)に到達する時点までを前記試料のゲル化時間として求め、予め求めたエンドトキシン濃度と該ゲル化時間との関係に基づいて前記二以上の被検液試料中のエンドトキシン濃度を並列して求める方法（時間測定法）を開発し、先に特許を取得している（特許文献1参照）。尚、この方法は、エンドトキシンに限らず、β―グルカン及びペプチドグリカンの測定にも使用し得るものであり、またゲル化時間でなく、色素が発生する活性化時間でも同様にして行うことができるものである。
【０００５】
【特許文献1】
特許第１８６０５７６号
【発明が解決しようとする課題】
上記時間測定法は、感度良く効率的に測定ができるという利点があったが、測定時間内にゲル化判定閾値に到達しない検体については、所定の濃度以下という結果しか得られなかった。このように一定の測定時間内で陽性判定されないサンプルに対しては、測定値が出せないため、測定時間を長くして再測定を実施するか、あらかじめ十分長い測定時間を設定して測定を実施する必要があった。そのため、処理能力の低下を生ずる問題があった。
【０００６】
この発明は、このような点に着目してなされたものであり、測定時間内に一定の閾値に到達しない検体について、測定時間内に得られた透過光若しくは吸光度の変化のデータを用いて一定の閾値に到達する時間を推定し、測定時間を延長することなく高感度測定を可能にする装置と方法を提供することを目的とする。
【０００７】
【課題を解決するための手段】
上記目的に沿う本発明のうち請求項１に記載の発明は、被検試料とゲル化若しくは色素発生試薬との混合後であって反応により変化する前の透過光量若しくは吸光度と、経時的に変化する透過光量若しくは吸光度との比率を測定する手段と、被検試料の反応による該比率の経時変化（タイムコース）をモニターする手段と、該比率が予め定めた所定の値(ゲル化若しくは色素発生判定閾値)に到達するまでの時間を測定する手段と、測定時間終了時点で前記ゲル化若しくは色素発生判定閾値に到達していない場合に、モニターしていたタイムコースを推定計算により測定時間以降に延長し推定タイムコース曲線を計算し、該推定タイムコース曲線が前記ゲル化若しくは色素発生判定閾値に到達する時間を算出する推定値計算プログラムと、を具備することを特徴とする。
【０００８】
ゲル化若しくは色素発生判定閾値は、被検試料とゲル化若しくは色素発生試薬との混合時点から前記比率が反応進行による変化を開始し始めた後の値であってかつこの値に達する時間と被検試料中の測定対象物質濃度との間に一定の関係が認められる予め設定された一定値とするのが良い（請求項２）。
【０００９】
ゲル化若しくは色素発生試薬としては、エンドトキシン、β―グルカン又はペプチドグリカン測定用試薬が好ましい（請求項３）。言いかえれば、測定対象物質としては、エンドトキシン、β―グルカン、ペプチドグリカン等が好ましい
【００１０】
本発明に使用するゲル化若しくは色素発生試薬は、カブトガニ血球由来の試薬（リムルステスト：エンドトキシン及びβ―グルカンの測定ができる試薬）、蚕体液由来の試薬（SLP試薬：β―グルカン及びペプチドグリカンの測定ができる試薬）又は合成基質含有試薬（エンドトキシン及びβ―グルカンの測定ができる試薬）であるのが好ましい（請求項４）。
【００１１】
本発明の測定値推定プログラムは、測定時間終了時点付近のタイムコースデータのノイズを補正し、その後の推定に使用できるレベルの補正タイムコースを得る手段を具備するのが好ましい（請求項５）。
【００１２】
本発明のプログラムは、タイムコースの最終測定点及び／又はその前の連続する数点の測定ポイント（推定処理ポイント）を認定する手段と、該推定処理ポイント各点の傾きの平均値を算出し、推定継続の可否判定を実施する手段と、該推定処理ポイントを使用して、関数式１及び関数式２、要すればその他の関数式によるカーブフィッティングを実施する手段と、前記ゲル化若しくは色素発生判定閾値と前記関数式１及び関数式２、要すればその他の関数式によるカーブフィッティングの曲線との交点を算出し推定値の候補とする手段と、該関数式１によるカーブフィッティングと関数式２によるカーブフィッティングと要すればその他の関数式のカーブフィッティングとの推定値候補値の比を算出し、その比の値を基準とする計算式で推定値を算出する手段と、を具備するようにするのが良い（請求項６）。
ここで用いられる関数式１、関数式２及びその他の関数式は、一次式、二次式又は三次式等を表し、統計学等において、関数をあてはめることにより得られるカーブフィッティングを実施するために用いられる式であり、好ましくは最小自乗法等により求められる式等が挙げられる。本発明の場合、上記関数式１が一次式であり、上記関数式２が二次式であることが好ましい（請求項７）。
【００１３】
前記一次式による推定値Ｔ１と前記二次式による推定値Ｔ２との差が、所定の値より小さい場合は推定値Ｔ１とし、所定の値より大きい場合は、推定値を（Ｔ１＋Ｔ２）／２とするのが良い（請求項８）。
【００１４】
ゲル化若しくは色素発生判定閾値に到達するまでの時間をゲル化時間若しくは活性化時間とし、予め求めた測定物質濃度と該ゲル化時間若しくは活性化時間との関係に基づいて被検試料中の測定物質濃度を求める手段を具備するのが良い（請求項９）。
【００１５】
本発明の測定方法は、被検試料とゲル化若しくは色素発生試薬との混合後であって反応により変化する前の透過光量若しくは吸光度と、経時的に変化する透過光量若しくは吸光度との比率を測定し、被検試料の反応による透過光量の比率若しくは吸光度の経時変化（タイムコース）をモニターし、該比率が予め定めた一定の値(ゲル化若しくは色素発生判定閾値)に到達するまでの時間を測定するか、若しくは測定時間終了時点で前記ゲル化若しくは色素発生判定閾値に到達していない場合に、モニターしていたタイムコースを推定計算により測定時間以降に延長して推定タイムコース曲線を計算し、該推定タイムコース曲線が前記ゲル化若しくは色素発生判定閾値に到達する時間を算出することを特徴とする（請求項１０）。
【００１６】
ゲル化若しくは色素発生判定閾値に到達するまでの時間をゲル化時間若しくは活性化時間とし、予め求めた測定物質濃度と該ゲル化時間若しくは活性化時間との関係に基づいて被検試料中の測定物質濃度を求めるのが良い（請求項１１）。
【００１７】
要するに本発明は、モニターしているタイムコースを測定時間以降に延長することにより、将来生じるゲル化時間若しくは色素発生時間（活性化時間）を推定するという従来全く行われていなかった方法を行うことによって、該推定した閾値に到達するまでの時間と一定の相関関係が認められる被検試料中の測定対象物質濃度を短時間で測定し得るようにしたことを要旨とするものである。
【００１８】
【発明の実施の形態】
次に、本発明の実施の形態を説明する。
【００１９】
測定対象物質（例えば、エンドトキシン、β―グルカン又はペプチドグリカン等）を含有する被検試料とＬＡＬ試薬若しくはSLP試薬とを混合して得た試液に対し、光源（例えばＬＥＤ）からの光線を照射し、試液を通過した光線は、光電検出器で透過光量に対応する電気量に変換される。
【００２０】
光電検出器で検知された透過光量Ｉ（ｔ）は、被検試料とＬＡＬ試薬若しくはSLP試薬とを混合した後の反応時間に対して、図１に示すような経時変化を示す。即ち、混合直後の透過光量Ｉｏから始まってゆっくりとした透過光量の変動を示すａの段階、次いで透過光量Ｉ（ｔ）が急激に減少するｂの段階、最後にＩ（ｔ）が殆ど変動のない一定値を示すｃの段階、以上３つの段階を経て試液はゲル化若しくは発色する。合成基質を用いた色素生成についても同様の変化を示すことが確認されている。
【００２１】
測定対象物質がエンドトキシンであってLAL試薬を用いた場合、上記ｃの段階に至った試液が従来の目視判定でゲル化反応陽性と判定される状態であること、更に図２に示すように、反応開始からｂの段階及びｃの段階へ到達するまでの所要時間が、被検試料中のエンドトキシン濃度に相関する。また、LAL試薬以外のSLP試薬及び合成基質含有試薬を用いた場合の色素発生についても同様の図になることは実験により確認されている。
【００２２】
光学系を複数準備し、複数の被検試料中の測定対象物質を測定する場合、測定されるＩ（ｔ）は、各光学系、各試料毎に全くばらばらの値を示すので、そのままでは各光学系を等価と考えて並列測定することはできない。本発明においては、各光学系のＩ（ｔ）を経時測定し、その反応初期段階（ａ）の透過光量と経時的に変化する透過光量Ｉ（ｔ）とｔの比Ｒ（ｔ）を求め、この値を用いて目的の測定を行うので、各光学系の感度差及び試料の違いによる感度差をキャンセルすることができる。
【００２３】
Ｒ（ｔ）は、図１におけるゲル化開始（色素発生）前のａの段階の透過光量（即ち、混合した直後の試液の揺動やノイズを除去した正しい初期透過光量）で光学系毎の特性を補正した透過光量の比率であり、各光学系で互換可能な量である。Ｒ（ｔ）の経時変化は図３に示すようにＩ（ｔ）と相似の挙動を示すが、本発明では、ゲル化進行（色素発生）の段階ｂにゲル化（色素発生）判定閾値Ｒｔｈを任意に設定し、反応開始からＲ（ｔ）がＲｔｈに到達するまでの時間を試料のゲル化時間Ｔ_G（活性化時間Ｔ_A）として検知する。
【００２４】
図２に示すように、エンドトキシン濃度とゲル化時間Ｔ_Gとは、それぞれ両対数目盛でプロットすると直線関係を示す。エンドトキシン濃度と活性化時間Ｔ_Aとも同様の直線関係を示す。従って、上記ゲル化時間Ｔ_G（活性化時間Ｔ_A）を検知することによって、エンドトキシン濃度を求めることができる。ゲル化（色素発生）判定閾値Ｒｔｈは、広範囲にわたってこのような直線関係が得られる値を選択する。
【００２５】
上記説明は、透過光量を測定して行ったが、吸光度を測定して行っても同様の結果が得られる。透過光量変化に基づいて目的物の定量を行う装置であれば、透過光量変化と吸光度変化は実質的に同義になるからである。
【００２６】
本発明は、測定終了時間ではＲｔｈに到達しない検体について、測定時間内に得られた透過光量若しくは吸光度の変化のデータを用いることにより、測定時間を延長することなく、一定のＲｔｈに到達する時間を推定するものである。
【００２７】
従って、本発明の装置は、前記公知の特許文献1記載の被検試料を収容する試料キューベットを恒温状態とする恒温槽と、該試料キューベットに光線を照射し、透過光量若しくは吸光度を検知する手段と、被検試料液の透過光量若しくは吸光度の測定開始を指示する手段と、測定開始時点からの経過時間を計時する手段とを具備した装置に加えて、透過光量若しくは吸光度の記憶手段（具体的にはメモリー）と未来の経時変化を推定し透過光量若しくは吸光度の比率あるいは吸光度が一定の値に到達するまでの時間を推定する手段（具体的には推定のプログラム）を付加したものである。
【００２８】
本発明方法は、測定時間終了時に一定の閾値に到達していない全ての検体について、各々推定処理を実行する。各々の検体では、モニターしていたタイムコースを推定計算により測定時間以降に延長し、推定タイムコース曲線を計算する。その推定タイムコース曲線が閾値に到達する時間を算出し、推定値とする。
【００２９】
実際の個々の被検試料の推定処理は、図４の推定処理のフローチャートに示す次のステップで実行する。
【００３０】
ノイズ除去のためタイムコースのスムージング（平滑化）を実行する。
タイムコースの最終測定点を含みその前の連続する数点の測定ポイント（該当ポイントをまとめて推定処理ポイントと呼び、その点数ｎは変更可能とする。図５にその説明図を示す）の各ポイントにおける傾き（微分係数）を算出する。
推定処理ポイント各点の傾きの平均値の算出と推定継続の判定を実施する。
推定処理ポイントを使用して、関数式１及び関数式２、更にその他の関数式によるカーブフィッティング、例えば一次式及び二次式によるカーブフィッティングを実行する。
以下、一次式及び二次式によるカーブフィッティングを用いる場合を例にとると、ゲル化（色素発生）判定閾値と4で求めた一次式及び二次式フィッティングの曲線との交点を算出し、推定値の候補とする。
二次式フィッティングと一次式フィッティングの推定値候補値の比を算出し、その比の値を基準とする計算式で推定値を算出する。
【００３１】
次に、各々のステップをより詳細に説明する。
ノイズ除去のためタイムコースのスムージング（平滑化）
推定処理を必要とする検体のタイムコースは、たいていの場合、その変化が緩やかなため、ノイズによる影響を受けやすい。次ステップ以降の処理の信頼性を上げるためには、ノイズを極力低減しておく必要がある。一般的な雑音軽減処理の移動平均処理でも良いが、後記本発明の実施例においては、種々の移動平均処理を実行できるSavitzky-Golay法（参考文献１）を利用し、重み付き移動平均処理を実施している。Savitzky-Golay法（参考文献１）：A Savitzky, M.J.E.Golay, Smoothing and Differentiation of Data by Simplified Least Squares Procediares, " Analytical Chemistry " Vol.36, No.8, pp. 1627-1639 (1964)。
【００３２】
これを要約すると、ある点における平滑化値は、その点を含む前後数点（一般的には、測定点を含む５〜１１点を使用する。）の測定点を用いて重み付き平均値を計算する。ランダムな雑音成分は、平均化により相殺され、実波形に近い成分が取り出せる。
【００３３】
推定処理ポイントの各ポイントにおける傾き（微分係数）の算出
測定打ち切り時間付近でのタイムコースの変化量を表わす指標として、測定終了付近の測定ポイント、すなわち推定処理ポイントの各ポイントで各々傾きを計算する。各ポイントの傾きを求めるために、その点の前後の測定点を利用して計算するが、その処理方法は公知の方法で行えば良い。傾きの計算においても、ノイズを軽減するため、平滑化処理が事前に必要となるので、後記実施例では、平滑化と傾きを求めるための処理である微分演算を移動平均処理の一環として実行できる前記参考文献１に記載の方法を利用し、平滑化微分処理を実施して各点の傾きを算出している。
【００３４】
推定処理ポイント各点の傾きの平均値の算出と推定継続の判定
リムルステストの反応には、見かけ上反応が進まない期間（反応ラグタイム）が存在する。測定打ち切り時間において、反応ラグタイムの最中である場合や、反応ラグタイム期間終了後間もなくタイムコースがほとんど変化していない場合、推定を実施することが困難であるか、推定値が非常に大きな値となり、十分な精度が得られない。またタイムコースが上昇している場合は、推定できない。
【００３５】
推定が可能かどうかの判定基準として、測定打ち切り時間付近での平均的なタイムコースの傾きを用いることとし、上記２．で求めた推定処理ポイント各点の傾きを計算する。この値が一定の基準値ｔより小さい時推定を打ち切ることとする。例えば、この基準値ｔとしては、経験的にタイムコースがほとんど減少していないと考えても良いとみなせる−0.0001（%/分）を使用することができる。これより大きい場合には減少していないかあるいは上昇していると見なし推定を打ち切る。
【００３６】
推定処理ポイントを使用する一次式及び二次式によるカーブフィッティング
推定処理ポイントのデータを使用して、一次式及び二次式の2種類のカーブフィッティングを実行する。カーブフィッティングとは、離散的にサンプリングした観測点（測定点）から、最小自乗法によって最も近いと思われる連続曲線を求めるための処理である。ここでは、曲線として一次多項式（一次式）と二次多項式（二次式）を使用する場合の２通りの方法を同時に実行する。２通りの式を使う理由は、完全に一致するような曲線を見つけることが困難であり、一次式を使った場合の推定結果と二次式を使った場合の推定結果の中間的な値が、推定値として最適であることが、実験により確認されているからである。一次式及び二次式のカーブを測定終了時間以降に延長したものが、タイムコースの推定曲線となる。
【００３７】
ゲル化（色素発生）判定閾値と4で求めた一次式及び二次式フィッティングの曲線との交点の算出
一次式及び二次式の各推定曲線とゲル化（色素発生）判定閾値との交点を算出する。これら2つの値が、推定タイムコースの測定値［ゲル化時間（活性化時間）］、すなわち推定測定値の候補となる。
【００３８】
二次式フィッティングと一次式フィッティングの推定値候補値の比を算出し、その比の値を基準とする計算式で推定値を算出する
タイムコースがほぼ直線と見なせる場合、一次式、二次式の推定結果はほとんど同じ値となる。一方、タイムコースの湾曲度が大きいほど、一次式と二次式の推定結果に差が出る。
【００３９】
ここで、二次式の推定値T_２と一次式の推定値T_１の比T_２／T_１が1に近いこと（すなわちT_１とT_２の値が近い）を示す指標として判定係数αを導入し、比が1±αの範囲に入る場合（すなわちT_１とT_２が十分近いとみなせる）場合には、一次式の値を推定値として採用する。それ以外の場合は、一次式と二次式の各推定値の中点を推定値として採用する。中点を採用する理由は、実験的に、一次式では交点が実際の値より少し大きくなり、二次式では交点が実際の値より少し小さくなり、一次式と二次式の交点の中間的な値が最適な値であることがわかっているからである。種々の実データから検討した結果、αとして好ましくは０．１〜０．３、後記実施例では0.2という値を使用している。
【００４０】
次に、実施例を挙げて、本発明を更に説明するが、本発明はこの実施例に限定されない。
【００４１】
【実施例】
【００４２】
実際に測定した種々の測定値を示すサンプルを多数集め、測定時間をソフトウェアの処理によって実際の測定時間より前の８０分で打ち切ることにより、強制的に推定処理を行わせ、その結果の性能を検証した。この実施例では、推定処理ポイントの点数ｎは5点、判定係数αとして0.2を使用した。得られた結果の推定時間と推定時間の実測時間に対するずれを次表１に示す。尚、比較のため実測時間も併記した。
【００４３】
【表１】

表１をグラフ化したものが、図６のグラフである。図６は、X軸には実測値、Y軸には推定値の実測値に対するずれを示す指標である計算式［（推定値―実測値）÷実測値］を割り当てたグラフである。
【００４４】
上記結果より明らかなように、本発明方法は、80分の測定で、150分位までなら±20％以内の精度で推定値を取得することが可能であり、十分実用的な性能である。
【００４５】
本発明によれば、所定の測定時間内に一定のゲル化判定閾値に到達していない検体について、過去のデータから未来の経時変化を推定して測定するものであるので、測定値が出せないとか、非常に長い測定時間を要する等の問題が解消し、迅速な測定が可能となる。
【００４６】
【発明の効果】
本発明によれば、測定時間内に一定の閾値に到達しない検体について、測定時間内に得られた透過光量若しくは吸光度の変化のデータを用いることにより測定時間を延長することなく一定の閾値に到達する時間を推定するというこの種方法では全く行われていなかった方法を行うことによって、長い測定時間を必要としたリムルステスト等の生化学的若しくは免疫学的測定を短時間で行うことができるという画期的な効果を奏する。
【図面の簡単な説明】
【図１】試料透過光量の経時変化を示す線図である。
【図２】エンドトキシン濃度とゲル化時間の検量関係を示す線図である。
【図３】試料透過光量比率の経時変化を示す線図である。
【図４】本発明の測定方法を示すフローチャートである。
【図５】ポイント数が５点の場合の推定処理ポイントの説明図である。
【図６】実施例で得たゲル化時間の実測値と推定値の実測値に対するずれを示すグラフである。[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a biochemical or immunological measuring device and a measuring method having a measured value estimating function used for measuring the concentration of a substance to be measured, and more specifically, does not reach a certain threshold value within a measuring time. For this reason, the present invention relates to a biochemical or immunological measuring apparatus and a measuring method having a function of estimating a measured value capable of quickly measuring a sample requiring a long time for measurement.
[0002]
[Prior art]
The present invention is mainly used for the measurement of endotoxin, β-glucan or peptidoglycan, and endotoxin which is a typical example thereof is a typical pyrogen. It is known that when the endotoxin-contaminated blood or injection solution is injected into a living body, serious side effects such as strong fever and shock are caused.
[0003]
In recent years, as a method for measuring endotoxin, it has been found that a limulus amoebocyte lysate (hereinafter, abbreviated as LAL) reacts with endotoxin to form a gel, and a method for measuring endotoxin using this is described. Is being developed.
[0004]
However, the conventional method for measuring endotoxin has a problem in that an accurate and stable measurement value is obtained. Therefore, the present applicant assigns each of two or more test samples, each of which is mixed with a gelling reagent, to each of light beams. And the ratio between the maximum value of the transmitted light amount after mixing (the value before the change by the reaction) and the transmitted light amount that changes with time is measured, and the ratio starts to rapidly decrease from the mixing time. Is determined as the gelation time of the sample until the time when the value reaches a predetermined value (gelation determination threshold value) and based on the relationship between the previously determined endotoxin concentration and the gelation time. Thus, a method (time measurement method) for obtaining the endotoxin concentrations in the two or more test liquid samples in parallel has been developed, and a patent has been previously obtained (see Patent Document 1). This method can be used not only for endotoxin, but also for measurement of β-glucan and peptidoglycan, and can be performed not only for gelation time but also for activation time when a dye is generated. It is.
[0005]
[Patent Document 1]
Patent No. 1860576 [Problems to be solved by the invention]
The above-described time measurement method has an advantage that measurement can be performed efficiently with high sensitivity, but for a sample that does not reach the gelation determination threshold within the measurement time, only a result of a concentration of a predetermined concentration or less was obtained. Since a measurement value cannot be output for a sample for which a positive determination is not made within a certain measurement time, either increase the measurement time and perform re-measurement, or set a sufficiently long measurement time in advance and perform the measurement. I needed to. Therefore, there is a problem that the processing capacity is reduced.
[0006]
The present invention has been made by paying attention to such a point, and for a sample that does not reach a certain threshold value within the measurement time, a constant value is obtained by using data of a change in transmitted light or absorbance obtained within the measurement time. It is an object of the present invention to provide an apparatus and a method for estimating a time required to reach a threshold value and performing high-sensitivity measurement without extending the measurement time.
[0007]
[Means for Solving the Problems]
The invention according to claim 1 of the present invention according to the above object is characterized in that the amount of transmitted light or absorbance after mixing of a test sample with a gelling or dye-forming reagent but before changing by a reaction changes with time. Means for measuring the ratio of the amount of transmitted light or absorbance to light, means for monitoring the change over time (time course) of the ratio due to the reaction of the test sample, and means for measuring the ratio of the ratio to a predetermined value (gelation or dye generation). (Measurement threshold) means to measure the time until reaching, and, when the gelation or dye development determination threshold is not reached at the end of the measurement time, the time course that was monitored by the estimation calculation after the measurement time An estimated value-course calculation program for calculating an estimated time-course curve by extending the estimated time-course curve and calculating a time required for the estimated time-course curve to reach the gelation or dye development determination threshold value. Characterized in that it.
[0008]
The gelation or dye generation determination threshold is a value after the ratio starts to change due to the progress of the reaction from the time of mixing of the test sample with the gelation or dye generation reagent, and the time required to reach this value and the threshold. It is preferable to use a preset constant value that has a certain relationship with the concentration of the substance to be measured in the test sample (claim 2).
[0009]
As the gelling or dye-forming reagent, a reagent for measuring endotoxin, β-glucan or peptidoglycan is preferable (claim 3). In other words, the substance to be measured is preferably endotoxin, β-glucan, peptidoglycan or the like.
The gelling or dye-forming reagent used in the present invention includes a horseshoe crab blood cell-derived reagent (limulus test: a reagent capable of measuring endotoxin and β-glucan) and a silkworm fluid-derived reagent (SLP reagent: measurement of β-glucan and peptidoglycan). (A reagent capable of measuring endotoxin and β-glucan) is preferable (claim 4).
[0011]
It is preferable that the measurement value estimation program of the present invention includes means for correcting noise of the time course data near the end of the measurement time and obtaining a corrected time course at a level usable for subsequent estimation.
[0012]
The program of the present invention is means for certifying the last measurement point of the time course and / or several consecutive measurement points (estimated processing points) before the time course, and calculates the average value of the slope of each of the estimated processing points. Means for determining whether or not continuation of estimation is possible, means for performing curve fitting based on the

function formulas

1 and 2 and, if necessary, other function formulas using the estimation processing points, Means for calculating an intersection between the occurrence determination threshold value and the curve of the curve fitting by the

function formulas

1 and 2 and, if necessary, other function formulas, and as a candidate for an estimated value; and a curve fitting by the function formula 1 and a function formula. 2 and, if necessary, the ratio of the estimated value candidate value to the curve fitting of other function formulas is calculated, and the ratio is estimated by a calculation formula based on the ratio value. It is good so as to comprise means for calculating a value, the (claim 6).
The function formula 1, the function formula 2, and other function formulas used here represent a linear formula, a quadratic formula, a cubic formula, and the like, and are used for performing curve fitting obtained by fitting a function in statistics or the like. This is an equation used, preferably an equation obtained by a least square method or the like. In the case of the present invention, it is preferable that the function formula 1 is a linear formula and the function formula 2 is a quadratic formula (claim 7).
[0013]
When the difference between the estimated value T1 based on the linear expression and the estimated value T2 based on the quadratic expression is smaller than a predetermined value, the estimated value is T1, and when the difference is larger than the predetermined value, the estimated value is (T1 + T2) / 2. (Claim 8).
[0014]
The time required to reach the gelation or dye generation determination threshold is defined as the gelation time or activation time, and the measurement in the test sample is performed based on the relationship between the measured substance concentration and the gelation time or activation time determined in advance. It is preferable to provide a means for determining the substance concentration (claim 9).
[0015]
The measurement method of the present invention measures the ratio of the transmitted light amount or absorbance after mixing of the test sample with the gelling or dye-forming reagent but before the change due to the reaction, and the transmitted light amount or absorbance that changes over time. Then, the ratio of the amount of transmitted light or the change over time in absorbance due to the reaction of the test sample (time course) is monitored, and the time until the ratio reaches a predetermined constant value (threshold for gelation or dye generation) is determined. When measuring, or when the gelation or dye generation determination threshold is not reached at the end of the measurement time, the monitored time course is extended to the measurement time or later by the estimation calculation to calculate the estimated time course curve. Calculating a time required for the estimated time-course curve to reach the gelation or pigment generation determination threshold (claim 10).
[0016]
The time required to reach the gelation or dye generation determination threshold is defined as the gelation time or activation time, and the measurement in the test sample is performed based on the relationship between the measured substance concentration and the gelation time or activation time determined in advance. Preferably, the substance concentration is determined (claim 11).
[0017]
In short, the present invention performs a method that has not been performed at all in the past by estimating a gelation time or a dye generation time (activation time) that will occur in the future by extending the monitored time course beyond the measurement time. Accordingly, the gist is that the concentration of the measurement target substance in the test sample having a certain correlation with the time until the estimated threshold value is reached can be measured in a short time.
[0018]
BEST MODE FOR CARRYING OUT THE INVENTION
Next, an embodiment of the present invention will be described.
[0019]
A test solution containing a test sample containing a substance to be measured (eg, endotoxin, β-glucan or peptidoglycan) and a LAL reagent or an SLP reagent is irradiated with light from a light source (eg, an LED), The light beam that has passed through the test solution is converted into an electric quantity corresponding to the amount of transmitted light by the photoelectric detector.
[0020]
The transmitted light amount I (t) detected by the photoelectric detector shows a temporal change as shown in FIG. 1 with respect to the reaction time after the test sample and the LAL reagent or the SLP reagent are mixed. That is, the stage a, which shows a slow change in the transmitted light amount starting from the transmitted light amount Io immediately after mixing, the stage b, in which the transmitted light amount I (t) sharply decreases, and finally, the I (t) almost changes. The test solution gels or develops a color through the three steps above, step c showing a non-constant value. It has been confirmed that a similar change is observed in the formation of a dye using a synthetic substrate.
[0021]
When the substance to be measured is endotoxin and the LAL reagent is used, the test solution that has reached the step c is in a state where the gelation reaction is determined to be positive by conventional visual determination, and further, as shown in FIG. 2, The time required from the start of the reaction to the steps b and c correlates with the endotoxin concentration in the test sample. Further, it has been confirmed by an experiment that the same figure is obtained for the generation of a dye when an SLP reagent other than the LAL reagent and a synthetic substrate-containing reagent are used.
[0022]
When a plurality of optical systems are prepared and a substance to be measured in a plurality of test samples is measured, the measured I (t) indicates a completely different value for each optical system and each sample. It is not possible to measure in parallel by considering the optical system as equivalent. In the present invention, I (t) of each optical system is measured with time, and the ratio R (t) of the transmitted light amount I (t) and the transmitted light amount I (t), which changes with time, in the initial stage (a) of the reaction is determined. Since the target measurement is performed using this value, it is possible to cancel the difference in sensitivity between the optical systems and the difference in sensitivity due to the difference between samples.
[0023]
R (t) is the transmitted light amount at the stage a before the start of gelation (generation of a dye) in FIG. 1 (that is, the correct initial transmitted light amount obtained by removing the fluctuation and noise of the test solution immediately after mixing) for each optical system. This is the ratio of the amount of transmitted light whose characteristics have been corrected, and is an amount compatible with each optical system. The time-dependent change of R (t) shows a behavior similar to I (t) as shown in FIG. 3, but in the present invention, the gelation (dye generation) determination threshold Rth the arbitrarily set, detects the time from the start of the reaction to R (t) reaches the Rth as gel time T _G of the sample (activation time T _a).
[0024]
As shown in FIG. 2, the endotoxin concentration and the gelation time _TG show a linear relationship when plotted on a logarithmic scale. Also endotoxin concentration and the activation time T _A shows a similar linear relationship. Therefore, by detecting the gel time T _G (activation time T _A), it is possible to determine the endotoxin concentration. As the gelation (dye generation) determination threshold value Rth, a value that provides such a linear relationship over a wide range is selected.
[0025]
Although the above description was made by measuring the amount of transmitted light, similar results can be obtained by measuring the absorbance. This is because the change in the amount of transmitted light and the change in absorbance have substantially the same meaning as long as the device performs the quantitative determination of the target object based on the change in the amount of transmitted light.
[0026]
According to the present invention, for a sample that does not reach Rth at the measurement end time, the time to reach a certain Rth without extending the measurement time by using data of the change in the transmitted light amount or absorbance obtained within the measurement time. Is estimated.
[0027]
Therefore, the apparatus of the present invention detects a transmitted light amount or an absorbance by irradiating a sample cuvette containing a test sample described in the above-mentioned patent document 1 with a constant temperature bath for keeping a sample at a constant temperature, and irradiating the sample cuvette with light rays. Means for instructing the start of measurement of the amount of transmitted light or absorbance of the test sample liquid, and a means for measuring the time elapsed from the start of measurement. (Specifically, a memory) and a means (specifically, an estimation program) for estimating a temporal change in the future and estimating a ratio of a transmitted light amount or an absorbance or a time until the absorbance reaches a certain value are added. is there.
[0028]
According to the method of the present invention, an estimation process is performed for all samples that have not reached a certain threshold at the end of the measurement time. For each sample, the monitored time course is extended by the estimated calculation to the time after the measurement time, and the estimated time course curve is calculated. The time required for the estimated time course curve to reach the threshold value is calculated and used as an estimated value.
[0029]
The actual estimation process of each test sample is executed in the next step shown in the flowchart of the estimation process in FIG.
[0030]
Performs time course smoothing (smoothing) to remove noise.
Each of several consecutive measurement points including the final measurement point of the time course and preceding the points (collectively referred to as estimation processing points, the number n of which can be changed; an explanatory diagram is shown in FIG. 5) Calculate the slope (differential coefficient) at the point.
The calculation of the average value of the inclination of each point of the estimation processing point and the determination of the continuation of the estimation are performed.
Using the estimated processing points, curve fitting by the

function formulas

1 and 2 and other function formulas, for example, curve fitting by the linear formula and the quadratic formula is executed.
Hereinafter, taking the case of using the curve fitting by the primary expression and the quadratic expression as an example, the intersection of the gelation (dye generation) judgment threshold value and the curve of the primary expression and the quadratic expression fitting obtained in 4 is calculated and estimated. Let it be a value candidate.
The ratio between the estimated value candidate value of the quadratic expression and the estimated value candidate value of the linear expression fitting is calculated, and the estimated value is calculated by a calculation formula based on the value of the ratio.
[0031]
Next, each step will be described in more detail.
Time course smoothing (smoothing) to remove noise
In most cases, the time course of a sample requiring an estimation process is susceptible to noise because the change is gradual. In order to increase the reliability of the processing after the next step, it is necessary to reduce noise as much as possible. Moving average processing of general noise reduction processing may be used. However, in an embodiment of the present invention, weighted moving average processing is performed by using the Savitzky-Golay method (Reference Document 1) capable of executing various moving average processing. We are implementing. Savitzky-Golay method (Reference 1): A Savitzky, MJEGolay, Smoothing and Differentiation of Data by Simplified Least Squares Procediares, "Analytical Chemistry" Vol. 36, No. 8, pp. 1627-1639 (1964).
[0032]
In summary, the smoothed value at a certain point is obtained by calculating a weighted average value using several measurement points before and after that point (generally, 5 to 11 points including the measurement point are used). calculate. The random noise component is canceled by the averaging, and a component close to the actual waveform can be extracted.
[0033]
Calculation of the slope (differential coefficient) at each point of the estimated processing point As an index indicating the amount of change in the time course near the measurement cutoff time, the slope is calculated at each measurement point near the end of the measurement, that is, at each point of the estimated processing point. . In order to obtain the inclination of each point, calculation is performed using measurement points before and after that point, and the processing method may be performed by a known method. Also in the calculation of the gradient, a smoothing process is required in advance to reduce noise, so in the later-described embodiment, a differential operation for obtaining the smoothing and the gradient can be executed as a part of the moving average process. Utilizing the method described in Reference 1, the slope of each point is calculated by performing a smoothing differentiation process.
[0034]
The reaction of the Limulus test for calculating the average value of the inclination of each point of the estimation processing point and determining the continuation of the estimation has a period during which the reaction does not seem to progress (reaction lag time). In the measurement censoring time, when the reaction lag time is in the middle, or when the time course does not substantially change shortly after the end of the reaction lag time period, it is difficult to perform the estimation, or the estimated value is very large. Value, and sufficient accuracy cannot be obtained. If the time course is rising, it cannot be estimated.
[0035]
As a criterion for determining whether or not estimation is possible, an average time course slope near the measurement cutoff time is used. Calculate the inclination of each point of the estimated processing point obtained in the above. When this value is smaller than a predetermined reference value t, the estimation is terminated. For example, as the reference value t, it is possible to use -0.0001 (% / min), which can be considered empirically that the time course has hardly decreased. If it is larger than this, it is regarded as not decreasing or rising, and the estimation is terminated.
[0036]
Two types of curve fitting, a primary equation and a quadratic equation, are performed using the data of the estimated processing points using the linear and quadratic equations using the estimated processing points. The curve fitting is a process for obtaining a continuous curve which is considered to be the closest from the discretely sampled observation points (measurement points) by the least squares method. Here, two methods in the case of using a first-order polynomial (first-order equation) and a second-order polynomial (second-order equation) as a curve are simultaneously executed. The reason for using the two formulas is that it is difficult to find a curve that completely matches, and the intermediate value between the estimation result using the linear expression and the estimation result using the quadratic expression is This is because it has been confirmed by experiments that the estimation value is optimal. A curve obtained by extending the curves of the primary expression and the secondary expression after the measurement end time becomes the estimated curve of the time course.
[0037]
Calculation of the intersection of the gelation (dye generation) judgment threshold with the curve of the primary equation and the quadratic equation determined in 4 The intersection of each estimated curve of the primary equation and the quadratic equation with the gelation (dye generation) judgment threshold calculate. These two values are the measured values of the estimated time course [gel time (activation time)], that is, candidates for the estimated measured value.
[0038]
Calculate the ratio between the estimated value candidate value of the quadratic fitting and the primary value fitting, and calculate the estimated value by the calculation formula based on the value of the ratio. Are almost the same. On the other hand, as the degree of curvature of the time course increases, the difference between the primary expression and the estimation result of the secondary expression increases.
[0039]
Here, the judgment coefficient α is used as an index indicating that the ratio T ₂ / T ₁ between the estimated value T ₂ of the quadratic expression and the estimated value T ₁ of the linear expression is close to ₁ (that is, the values of T ₁ and T ₂ are close). introducing, when entering the range of ratios is 1 ± alpha (regarded as ie T ₁ and T ₂ are close enough), then, to adopt the value of a primary expression as an estimate. In other cases, the midpoint between the estimated values of the linear expression and the quadratic expression is adopted as the estimated value. The reason for adopting the midpoint is that, experimentally, the intersection in the linear equation is slightly larger than the actual value, the intersection in the quadratic equation is slightly smaller than the actual value, and the intermediate point between the intersection of the primary equation and the quadratic equation. This is because it is known that the optimal value is the optimal value. As a result of an examination based on various actual data, α is preferably 0.1 to 0.3, and 0.2 is used in Examples described later.
[0040]
Next, the present invention will be further described with reference to examples, but the present invention is not limited to these examples.
[0041]
【Example】
[0042]
A large number of samples showing various actually measured values are collected, and the measurement time is cut off by software processing at 80 minutes before the actual measurement time, thereby forcibly performing the estimation processing, and the performance of the result is reduced. Verified. In this embodiment, the number n of estimated processing points is 5, and the determination coefficient α is 0.2. Table 1 shows the estimated time of the obtained result and the difference between the estimated time and the actually measured time. The actual measurement time is also shown for comparison.
[0043]
[Table 1]

FIG. 6 is a graph of Table 1. FIG. 6 is a graph in which the X-axis is assigned a measured value, and the Y-axis is assigned a calculation formula [(estimated value−actual value) ÷ actual value] which is an index indicating a deviation of the estimated value from the actual value.
[0044]
As is clear from the above results, the method of the present invention is capable of obtaining an estimated value with an accuracy of ± 20% within a range of up to 150 minutes in a measurement of 80 minutes, which is a sufficiently practical performance.
[0045]
According to the present invention, a sample that does not reach a certain gelation determination threshold within a predetermined measurement time is measured by estimating a future change with time from past data. And the problem that an extremely long measurement time is required is solved, and quick measurement is possible.
[0046]
【The invention's effect】
According to the present invention, for a sample that does not reach a certain threshold value within the measurement time, it reaches a certain threshold value without extending the measurement time by using data of the change in the amount of transmitted light or absorbance obtained within the measurement time. By performing a method that has not been performed at all with this kind of method of estimating the time to perform, it is possible to perform a biochemical or immunological measurement such as a Limulus test that requires a long measurement time in a short time. It has a periodical effect.
[Brief description of the drawings]
FIG. 1 is a diagram showing a change over time of a sample transmitted light amount.
FIG. 2 is a diagram showing a calibration relationship between endotoxin concentration and gelation time.
FIG. 3 is a diagram showing a temporal change of a sample transmitted light amount ratio.
FIG. 4 is a flowchart showing a measuring method of the present invention.
FIG. 5 is an explanatory diagram of estimation processing points when the number of points is five.
FIG. 6 is a graph showing the difference between the actually measured gelling time and the estimated gelling time from the actually measured value obtained in the example.

Claims

Means for measuring the ratio of the transmitted light amount or absorbance after mixing of the test sample with the gelling or dye-forming reagent but before the change due to the reaction, and the transmitted light amount or absorbance that changes over time; Means for monitoring the change over time (time course) of the ratio due to the reaction of the above, means for measuring the time until the ratio reaches a predetermined value (threshold for judging gelation or dye formation), and measuring time When the gelation or dye generation determination threshold has not been reached at the end point, the monitored time course is extended by the estimation calculation after the measurement time to calculate an estimated time course curve, and the estimated time course curve is An estimated value calculation program for calculating a time to reach a gelation or dye generation determination threshold, and a biochemical having a measured value estimation function, Properly immunological measurement device.

The gelation or dye generation determination threshold is a value after the ratio starts to change due to the reaction progress from the time of mixing the test sample and the gelation or dye generation reagent, and the time to reach this value. 2. The measuring apparatus according to claim 1, wherein the measuring apparatus is a preset constant value at which a certain correlation is recognized between the concentration of the substance to be measured in the test sample.

3. The measuring device according to claim 1, wherein the gelling or dye-forming reagent is a reagent for measuring endotoxin, β-glucan or peptidoglycan.

The measuring device according to any one of claims 1 to 3, wherein the gelling or dye-forming reagent is a horseshoe crab blood cell-derived reagent, a silkworm body fluid-derived reagent, or a synthetic substrate-containing reagent for measuring endotoxin or β-glucan.

5. The measuring apparatus according to claim 1, wherein the program corrects noise of the time course data near the end of the measurement time, and obtains a corrected time course at a level usable for subsequent estimation.

The program is means for certifying the last measurement point of the time course and / or several consecutive measurement points (estimated processing points) before the time course, and calculating the average value of the slope of each of the estimated processing points, Means for determining whether continuation is possible, means for performing curve fitting using the function points 1 and 2 using the estimated processing points, the threshold for determining gelation or pigment generation, the function 1 and the function Means for calculating an intersection with the curve of the curve fitting according to equation 2 to be a candidate for an estimated value; And a means for calculating an estimated value using a calculation formula based on the value of the ratio.

The measuring device according to claim 6, wherein the function formula 1 is a linear formula, and the function formula 2 is a quadratic formula.

When the difference between the estimated value T1 based on the linear expression and the estimated value T2 based on the quadratic expression is smaller than a predetermined value, the estimated value is T1, and when the difference is larger than the predetermined value, the estimated value is (T1 + T2) / 2. The measuring device according to claim 7, wherein

The time required to reach the gelation or dye development determination threshold is defined as a gelation time or an activation time, and the test substance concentration in the test sample is determined based on a relationship between the measured substance concentration and the gelation time or the activation time. The measuring device according to any one of claims 1 to 8, further comprising a means for determining a concentration of the substance to be measured.

After mixing the test sample with the gelling or dye-forming reagent and before changing by the reaction, measure the ratio of the transmitted light amount or the absorbance that changes over time to the transmitted light amount or the absorbance that changes over time, and perform the reaction of the test sample. Monitoring the time-dependent change in the ratio of the amount of transmitted light or the absorbance (time course), and measuring the time until the ratio reaches a predetermined constant value (threshold for determination of gelation or dye formation) or the measurement time If the gelation or dye generation determination threshold has not been reached at the end point, the monitored time course is extended by the estimation calculation to the time after the measurement time to calculate an estimated time course curve, and the estimated time course curve is calculated. A biochemical or immunological measurement method, comprising calculating a time required to reach the gelation or pigmentation determination threshold.

The time required to reach the gelation or dye generation determination threshold is defined as a gelation time or an activation time, and the time in the test sample is determined based on the relationship between the concentration of the substance to be measured and the gelation time or the activation time determined in advance. The method according to claim 10, wherein the concentration of the test substance is determined.