Real-time sample shortage and suction failure detection

Abstract

A method for real-time detection of a sample shortage aspiration fault in an automated diagnostic analysis system includes spectral analysis of a pressure gradient waveform based on an aspiration pressure measurement signal. The spectral analysis can include identifying distinct transient behaviors in the pressure gradient waveform using a moving average filter or a wavelet transform, such as the continuous wavelet transform (CWT) or discrete wavelet transform (DWT). These methods accurately identify a sample shortage aspiration fault, allowing the automated diagnostic analysis system to timely terminate the detected sample shortage analysis to avoid potentially erroneous sample test results. A real-time detection device for a sample shortage aspiration fault is also provided, as well as other aspects.

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit of U.S. Provisional Patent Application No. 63 / 221,453, entitled "REAL-TIME SHORT SAMPLE ASPIRATION FAULT DETECTION," filed July 13, 2021, the disclosure of which is incorporated herein by reference in its entirety for all purposes.

[0002] The present disclosure relates to the aspiration of liquids in automated diagnostic analysis systems. [Background technology]

[0003] In medical testing, automated diagnostic analyzer systems can be used to analyze biological samples to identify analytes or other components in the sample. The biological sample can be, for example, urine, whole blood, serum, plasma, interstitial fluid, cerebrospinal fluid, etc. Such biological fluid samples are typically contained in sample containers (e.g., test tubes, vials, etc.) and can be transported to and from various imaging, processing, and analyzer stations within the automated diagnostic analyzer system via container carriers and automated trucks.

[0004] Automated diagnostic analytical systems typically include automated aspirating and dispensing devices configured to aspirate (draw) liquid from a liquid container (e.g., a sample of a biological liquid, or a liquid reagent, acid, or base to be mixed with the sample) and dispense the liquid into a reaction vessel (e.g., a cuvette). The aspirating and dispensing device typically includes a probe (e.g., a pipette) attached to a movable robotic arm or other automated mechanism that performs the aspirating and dispensing functions and transfers the sample or reagent to the reaction vessel.

[0005] During the aspiration process, a movable robotic arm, which can be controlled by a system controller or processor, can position a probe above a liquid container and then lower the probe into the container until the probe is partially submerged in the liquid. A pump or other suction device is then activated to aspirate (draw) a portion of the liquid from the container into the probe. The probe is then withdrawn from the container so that the liquid can be transferred to and dispensed into a reaction vessel for processing and / or analysis. During or after aspiration, the aspiration pressure signal can be analyzed to determine whether any abnormalities have occurred, such as aspirating an insufficient amount of liquid (hereinafter sometimes referred to as a short-sample aspiration failure). Summary of the Invention [Problem to be solved by the invention]

[0006] While conventional systems may be able to detect some sample shortage aspiration failures, such detection may be inaccurate, may be computationally expensive, and / or may occur too late in the sample analysis process to prevent sample shortages from adversely affecting test results.

[0007] Therefore, there is a need for improved methods and devices for accurate real-time detection of under-sample aspiration faults to avoid erroneous or inaccurate sample testing. [Means for solving the problem]

[0008] Some embodiments provide a method for detecting a sample-deficient aspiration fault in an automated diagnostic analyzer system. The method includes performing an aspiration pressure measurement via a pressure sensor in the automated diagnostic analyzer system as a liquid is being aspirated. The method further includes analyzing the aspiration pressure measurement signal waveform via a processor executing an algorithm. The algorithm is configured to derive a gradient waveform from the aspiration pressure measurement signal waveform and calculate a moving average of the gradient waveform or calculate a wavelet transform of the gradient waveform. The method further includes identifying and responding to a sample-deficient aspiration fault via the processor in response to the analyzing.

[0009] In some embodiments, an automated aspirating and dispensing device is provided that includes a robotic arm, a probe coupled to the robotic arm, a pump coupled to the probe, a pressure sensor configured to perform aspiration pressure measurements as liquid is aspirated through the probe, and a processor configured to execute an algorithm to detect and respond to low-sample aspiration faults during the aspiration process, wherein the algorithm is configured to analyze an aspiration pressure measurement signal waveform received from the pressure sensor by deriving a gradient waveform from the aspiration pressure measurement signal waveform and performing a spectral analysis of the gradient waveform by calculating a moving average or wavelet transform of the gradient waveform.

[0010] In some embodiments, the non-transitory computer-readable storage medium includes a processor-executable algorithm configured to detect an under-sample aspiration fault based on a spectral analysis of a pressure gradient waveform derived from an aspiration pressure measurement signal waveform, the algorithm being configured to perform the spectral analysis of the pressure gradient waveform by calculating a moving average or a wavelet transform of the pressure gradient waveform.

[0011] Some embodiments provide a method for detecting a sample-deficient aspiration fault in an automated diagnostic analyzer system. The method includes deriving an aspiration pressure measurement signal waveform from aspiration pressure measurements taken by a pressure sensor while a liquid is being aspirated in the automated diagnostic analyzer system. The method also includes identifying a pattern in one or more first aspiration pressure measurement signal waveforms of successful aspiration and defining a time-windowed localization of an anomaly identified in one or more second aspiration pressure measurement signal waveforms, the anomaly being caused by a sample-deficient aspiration fault. The method further includes deriving a suitable discrimination metric for detecting the anomaly, wherein a simple thresholding, an unsupervised classifier, or a supervised learning-based classifier is used in conjunction with the discrimination metric to identify the anomaly in subsequent aspiration pressure measurement signal waveforms.

[0012] Further aspects, features, and advantages of the present disclosure will be readily apparent from the following detailed description and illustrations of numerous embodiments and example implementations, including the best mode contemplated for carrying out the invention. The present disclosure is also capable of other different embodiments, and its several details may be modified in various respects, all without departing from the scope of the present invention. For example, although the following description is directed to an automated diagnostic analysis system, the sample shortage aspiration fault detection method and apparatus disclosed herein can be readily applied to other automated systems that would benefit from accurate, real-time detection of sample shortage aspiration faults. The present disclosure is intended to cover all modifications, equivalents, and alternatives that fall within the scope of the appended claims (see further below).

[0013] The drawings described below are for illustrative purposes and are not necessarily drawn to scale. Accordingly, the drawings and descriptions are to be regarded as illustrative in nature and not as restrictive. The drawings are not intended to limit the scope of the invention in any way. [Brief explanation of the drawings]

[0014] [Figure 1] FIG. 1 is a top schematic view of an automated diagnostic analysis system configured to analyze a biological sample according to embodiments provided herein. [Figure 2] 2A and 2B are front views of sample containers according to embodiments provided herein. [Figure 3] 1 is a front schematic view of an aspirating and dispensing device according to embodiments provided herein. FIG. [Figure 4] 1 is a flowchart of a method for detecting a sample shortage aspiration fault in an automated diagnostic analyzer system according to embodiments provided herein. [Figure 5] 5A and 5B are graphs of an aspiration pressure signal waveform versus time for normal aspiration, according to embodiments provided herein, and an aspiration pressure gradient waveform versus time for the aspiration pressure signal waveform of FIG. 5A, according to embodiments provided herein. [Figure 6] 6A and 6B are graphs of an aspiration pressure signal waveform versus time for an abnormal aspiration (low sample aspiration failure) according to embodiments provided herein. 6B are graphs of an aspiration pressure gradient waveform versus time for the aspiration pressure signal waveform of FIG. 6A according to embodiments provided herein. [Figure 7] 7A and 7B are graphs of the moving average (MA) of a gradient waveform representing normal aspiration, according to embodiments provided herein, and a delta signal based on the MA of FIG. 7A, according to embodiments provided herein. [Figure 8] 8A and 8B are graphs of the moving average (MA) of a gradient waveform representing abnormal aspiration, according to embodiments provided herein, and a delta signal based on the MA of FIG. 8A, according to embodiments provided herein. [Figure 9] 10 is a graph of multiple delta signals based on moving averages (MA) of eight sample aspiration pressure signal waveforms according to embodiments provided herein. [Figure 10] 10 is a bar graph of signal-to-noise ratio (SNR) versus SNR metric for the delta signal of FIG. 9 of an 8-sample aspiration pressure signal waveform in accordance with embodiments provided herein. [Figure 11] 10 is a graph of pressure gradient versus time for eight test sample aspiration pressure signal waveforms according to embodiments provided herein. [Figure 12] 12 is a graph of a two-cluster classification of the maximum slope metric of the eight test sample aspiration pressure signal waveforms of FIG. 11 in accordance with embodiments provided herein. [Figure 13] 1 is a graph of SNR metric versus time for a sample of normal aspiration based on continuous wavelet transform (CWT), according to embodiments provided herein. [Figure 14] 10 is a graph of SNR metric versus time for samples of abnormal aspiration based on CWT, according to embodiments provided herein. [Figure 15] 1 is a graph of a sample of normal aspiration SNR metric versus time based on a discrete wavelet transform (DWT) according to embodiments provided herein. [Figure 16] 10 is a graph of SNR metric versus time for samples of abnormal aspiration based on DWT, according to embodiments provided herein. [Figure 17] 1 is a schematic diagram of a multi-level discrete wavelet transform (DWT) filter bank according to embodiments provided herein. DETAILED DESCRIPTION OF THE INVENTION

[0015] The embodiments described herein provide methods and devices for timely and accurate detection of under-sample aspiration faults in real time. A under-sample aspiration fault occurs when an aspiration does not aspirate a sufficient amount of liquid, which may be caused, for example, by the liquid container not containing enough liquid, a blockage, and / or an instrument defect (e.g., a defective aspiration pump, improper positioning of the probe in the liquid container due to a defective robotic arm, a software defect, etc.). In some embodiments, under-sample may be considered as less than 92 μL for a nominal / target aspiration volume of 100 μL. Other volumes may also be considered under-sample. Timely and accurate real-time detection of under-sample aspiration faults can enable an automated diagnostic analysis system to terminate the analysis of the sample and / or implement suitable error condition procedures to advantageously avoid erroneous analytical results. In some embodiments, detection of under-sample aspiration faults can be considered timely if detected during the aspiration process or shortly after completion of the aspiration process.

[0016] According to one or more embodiments, timely and accurate real-time detection of sample shortage aspiration failures can be implemented via a software or firmware algorithm running on a system controller, processor, or other similar computing device of an automated diagnostic analysis system or automated aspiration and dispensing device. In some embodiments, the algorithm can be a learning-based AI (artificial intelligence) algorithm. The algorithm can be configured to perform spectral analysis of a pressure gradient waveform derived from an aspiration pressure measurement signal provided by a pressure sensor to identify distinct transient behavior in the pressure gradient waveform. Spectral analysis can include time-domain analysis, such as moving average filter analysis or analysis of bandpass filtered signals, or the spectral analysis can use short-time fast Fourier transform (STFT) or wavelet transform analysis. Preferred embodiments can include moving average filter analysis for its simplicity, or wavelet transform analysis for its ability to localize across time and scale (spectral content).

[0017] The moving average filter analysis may include calculating the difference between a moving average (determined over a suitable moving average window) at suitable time steps within a suitable detection window of the gradient waveform during the aspiration process and the gradient waveform, after which a suitable signal-to-noise ratio (SNR) threshold may be applied to the calculated difference to classify the aspiration as normal or abnormal.

[0018] The signal discrimination metric may be one or more of the following: (a) some statistical measure such as the maximum, median, standard deviation, 75th percentile, etc. of the pressure gradient over a predetermined time window of interest, or (b) the difference between the pressure gradient over a predetermined time window of interest and a moving average of the pressure gradient.

[0019] Wavelet transform analysis can include using a continuous wavelet transform (CWT) or a discrete wavelet transform (DWT), interrogating a metric based on the transform coefficients at a particular range of scales, and then applying a suitable SNR threshold to classify the aspiration as normal or abnormal.

[0020] More advanced learning-based classifiers can be used to automatically set thresholds or discrimination boundaries that separate abnormal aspirates (under-specimen aspirates) from normal aspirates in a robust manner to achieve high classification accuracy. In some embodiments, k-means clustering can be used to classify aspirates as normal or abnormal based on an SNR metric in an unsupervised manner. Simple threshold-based fuzzy classifiers or simple rule-based criteria or schemas can alternatively be used to classify samples as normal or abnormal (under-specimen aspirates). Supervised learning-based classifiers, such as logistic function classifiers or support vector machines, can also be used.

[0021] Advantageously, both the moving average filter analysis and the wavelet transform analysis can be performed online and in real time during the aspiration process. Each online analysis has low computational complexity (O(N)) and low memory requirements, and can therefore be easily implemented in firmware or software.

[0022] In accordance with one or more embodiments, methods and apparatus for timely and accurate real-time detection of sample-deficient aspiration faults are described in more detail below with reference to FIGS.

[0023] FIG. 1 illustrates an automated diagnostic analyzer system 100 according to one or more embodiments. The automated diagnostic analyzer system 100 can be configured to automatically process and / or analyze biological samples contained in sample containers 102. The sample containers 102 can be received in the system 100 in one or more racks 104 located in a loading area 106. The loading area 106 can include a robotic container handler 108 that picks up a sample container 102 from one of the racks 104 and loads the sample container 102 into a container carrier 110 located on an automated track 112. The sample containers 102 can be transported via the automated track 112 throughout the system 100, for example, to a quality check station 114, an aspirating and dispensing station 116, and / or one or more analyzer stations 118A-C.

[0024] The quality check station 114 can pre-screen the biological sample for interfering substances or other undesirable characteristics to determine whether the sample is suitable for analysis. After passing pre-screening, the biological liquid sample can be mixed with a liquid reagent, acid, base, or other solution at the aspirating and dispensing station 116 to enable and / or facilitate analysis of the sample at one or more analyzer stations 118A-C. The analyzer stations 118A-C can analyze the sample for the presence, quantity, or functional activity of a target entity (analyte), such as DNA or RNA. Other analytes commonly tested for include enzymes, substrates, electrolytes, specific proteins, drugs of abuse, and therapeutic drugs. More or fewer analyzer stations 118A-C can be used in the system 100, and the system 100 can include other stations (not shown).

[0025] The automated diagnostic analysis system 100 may also include a computer 120, or alternatively, may be configured to remotely communicate with an external computer 120. The computer 120 may be, for example, a system controller or the like, and may have a microprocessor-based central processing unit (CPU) and / or other suitable computer processor. The computer 120 may include suitable memory, software, electronics, and / or device drivers for operating and / or controlling various components of the system 100 (including the quality check station 114, the aspirating and dispensing station 116, and the analyzer stations 118A-C). For example, the computer 120 may control the movement of the carrier 110 to and from the loading area 106, around the track 112, between the quality check station 114, the aspirating and dispensing station 116, and the analyzer stations 118A-C, as well as between other stations and / or components of the system 100. Quality check station 114, aspirate and dispense station 116, and one or more of analyzer stations 118A-C may be directly coupled to computer 120 or may communicate with computer 120 via a network 122, such as a local area network (LAN), a wide area network (WAN), or other suitable communications network, including wired and wireless networks. Computer 120 may be housed as part of system 100 or may be remote from system 100.

[0026] In some embodiments, the computer 120 can be coupled to a laboratory information system (LIS) database 124. The LIS database 124 can include, for example, patient information, tests to be performed on the biological sample, the date and time the biological sample was obtained, medical facility information, and / or tracking and routing information. Other information can also be included.

[0027] The computer 120 can be coupled to a computer interface module (CIM) 126. The CIM 126 and / or the computer 120 can be coupled to a display 128, which can include a graphical user interface. The CIM 126, in conjunction with the display 128, allows a user to access various control and status screens and input data into the computer 120. These control and status screens can display and enable control of some or all aspects of the quality check station 114, the aspirate and dispense station 116, and the analyzer stations 118A-C, which prescreen, prepare, and analyze biological samples in the sample containers 102. The CIM 126 can be used to facilitate interaction between a user and the system 100. The display 128 can be used to display menus, including icons, scroll bars, boxes, and buttons, through which a user (e.g., a system operator) can interface with the system 100. The menus can include a number of functional elements programmed to display and / or operate functional aspects of the system 100.

[0028] 2A and 2B show sample containers 202A and 202B, respectively, each representing the sample container 102 of FIG. 1. Sample containers 202A and 202B can be any suitable liquid container, including transparent or translucent containers, such as blood collection tubes, test tubes, sample cups, cuvettes, or other containers capable of containing a biological sample therein and allowing the biological sample to be prescreened, processed (e.g., aspirated), and analyzed. As shown in FIG. 2A, sample container 202A can include tube 230A and cap 232A. Tube 230A can include label 234A thereon, which can indicate patient, sample, and / or test information in the form of a bar code, alphanumeric characters, numeric characters, or a combination thereof. Tube 230A can contain biological sample 236A therein, which can include serum or plasma portion 236SP, sedimented blood portion 236SB, and gel separator 216GA positioned therebetween. As shown in FIG. 2B, sample vessel 202B, which may be structurally identical to sample vessel 202A, can contain a homogenous biological sample 236B therein, with tube 230B having a gel bottom 236GB.

[0029] 3 illustrates an aspirate / dispense device 316 according to one or more embodiments. The aspirate / dispense device 316 may be part of or represent the aspirate / dispense station 116 of the automated diagnostic analyzer system 100. It should be noted that the methods and devices described herein for detecting a sample shortage aspiration fault may be used with other embodiments of the aspirate / dispense device, such as those located in the analyzers 118A, 118B, and / or 118C.

[0030] The aspirator / dispenser 316 can aspirate and dispense biological samples (e.g., samples 236A and / or 236B), reagents, etc., into reaction vessels to enable or facilitate analysis of the biological samples at one or more analyzer stations 118A-118C. The aspirator / dispenser 316 can include a robot 338 configured to move a probe assembly 340 within the aspirator / dispenser station. The probe assembly 340 can include a probe 340P configured to aspirate a reagent 342, for example, from a reagent packet 344, as shown. The probe assembly 340 can also be configured to aspirate a biological sample 336 from a sample container 302 located at the aspirator / dispenser 316 (after its cap has been removed, as shown), for example, via the automated track 112. Reagents 342, other reagents, and portions of the sample 336 can be dispensed by the probe 340P into a reaction vessel, such as a cuvette 346. In some embodiments, the cuvette 346 can be configured to hold only a few microliters of liquid. Another portion of the biological sample 336 can be dispensed by probe 340P into another cuvette (not shown) along with other reagents or liquids.

[0031] The operation of some or all components of the aspirating and dispensing apparatus 316 can be controlled by a computer 320. The computer 320 can include a processor 320P and a memory 320M. The memory 320M can store software 320S executable by the processor 320P. The software 320S can include algorithms for controlling and / or monitoring the positioning of the probe assembly 340 and the aspirating and dispensing of liquid by the probe assembly 340. The software 320S can also include an algorithm 320A configured to detect a sample shortage aspiration fault, as described further below. In some embodiments, the algorithm 320A can be an artificial intelligence (AI) algorithm. The computer 320 can be a separate computing / control device coupled to the computer 120 (the system controller). In some embodiments, the configuration and functionality of the computer 320 can be implemented in and executed by the computer 120. Also, in some embodiments, the probe assembly positioning and / or probe assembly aspirating / dispensing functions can be implemented in a separate computing / control device or in the computer 120.

[0032] The robot 338 may include one or more robotic arms 342, a first motor 344, and a second motor 346 configured to move a probe assembly 340, for example, within the aspirating and dispensing station 116 of the system 100. The robotic arm 342 may be coupled to the probe assembly 340 and the first motor 344. The first motor 344 may be controlled by the computer 320 to move the robotic arm 342, and therefore the probe assembly 340, to a position above a liquid container. The second motor 346 may be coupled to the robotic arm 342 and the probe assembly 340. The second motor 346 may also be controlled by the computer 320 to move the probe 340P vertically into and out of a liquid container to aspirate or dispense liquid therefrom. In some embodiments, the robot 338 may also include one or more sensors 348, such as, for example, vibration, current or voltage, and / or position sensors, coupled to the computer 320 to provide feedback and / or facilitate the operation of the robot 338.

[0033] The aspirating-dispensing device 316 can also include a pump 350 that is mechanically coupled to the conduit 352 and controlled by the computer 320. The pump 350 can generate a vacuum or negative pressure (e.g., aspiration pressure) in the conduit 352 to aspirate liquid, and can generate a positive pressure (e.g., dispense pressure) in the conduit 352 to dispense liquid.

[0034] Aspirating-dispensing device 316 may further include a pressure sensor 354 configured to measure the aspiration and dispense pressure in conduit 352 and generate pressure data accordingly. The pressure data may be received by computer 320 and used to control pump 350. An aspiration pressure measurement signal waveform (versus time) may be derived by computer 320 from the received pressure data and input to algorithm 320A to detect sample-deficient aspiration faults in probe assembly 340 during the aspiration process. In embodiments where algorithm 320A is an AI algorithm, the aspiration pressure measurement signal waveform derived from the received pressure data from pressure sensor 354 may also be used to train the AI ​​algorithm to detect sample-deficient aspiration faults.

[0035] 4 illustrates a method 400 for detecting a sample-deficient aspiration fault in an automated diagnostic analyzer system, according to one or more embodiments. At process block 402, method 400 can begin by performing an aspiration pressure measurement via a pressure sensor in the automated diagnostic analyzer system as liquid is being aspirated. For example, the aspiration pressure measurement can be performed by pressure sensor 354 of aspirator / dispense device 316 (of FIG. 3), which can be part of aspirator / dispense station 116 of automated diagnostic analyzer system 100 (of FIG. 1).

[0036] At process block 404, the method 400 may include analyzing the aspiration pressure measurement signal waveform via a processor executing an algorithm configured to derive a gradient waveform from the aspiration pressure measurement signal waveform and calculate a moving average of the gradient waveform or a wavelet transform of the gradient waveform.

[0037] Analyzing the aspiration pressure measurement signal waveform to detect a sample shortage aspiration fault is based on the distinct transient behavior difference between the pressure measurement signal waveform of normal aspiration and the pressure measurement signal waveform of abnormal aspiration (indicative of a sample shortage aspiration fault).

[0038] Figure 5A shows a graph 500A of an aspiration pressure signal waveform versus time for a normal aspiration, according to one or more embodiments, and Figure 5B shows a graph 500B of the slope waveform (i.e., the time derivative of the pressure signal waveform, d(pressure signal) / dt) versus time of the aspiration pressure signal waveform of Figure 5A, according to one or more embodiments. In contrast, Figure 6A shows a graph 600A of an aspiration pressure signal waveform versus time for an abnormal (low-sample) aspiration, according to one or more embodiments, and Figure 6B shows a graph 600B of the slope waveform (i.e., the time derivative of the pressure signal waveform, d(pressure signal) / dt) versus time of the aspiration pressure signal waveform of Figure 6A, according to one or more embodiments. The aspiration pressure signal waveforms of graphs 500A and 600A may each have been generated by an aspirator-dispensing device, such as aspirator-dispensing device 316 (of FIG. 3), and the slope waveforms of graphs 500B and 600B may each have been derived by algorithm 320A from the aspiration pressure signal waveforms of graphs 500A and 600A, respectively. Note the circled portions 602A and 602B of graphs 600A and 600B, respectively, compared to the corresponding portions in graphs 500A and 500B, respectively. These differences are detectable by method 400.

[0039] In some embodiments, after deriving the gradient waveform, process block 404 further includes analyzing the derived gradient waveform via a processor executing an algorithm by calculating a moving average of the gradient waveform and then calculating the difference between the moving average and the gradient waveform at suitable time increments (e.g., every 10 milliseconds) during the aspiration process. These calculated differences may be referred to as the delta signal. The moving average may, in some embodiments, be based on a moving average window of approximately 10 milliseconds (±10%). The analysis performed in process block 404 may continue by calculating the noise floor amplitude, which in some embodiments may be the root-mean-square (RMS) amplitude of the delta signal from t=0 to 150 milliseconds of the aspiration process. One or more signal amplitude metrics (e.g., absolute average, RMS, or 75th percentile) of the delta signal may be calculated over a detection window, which in some embodiments may be 270 to 320 milliseconds of the aspiration process. The SNR may then be calculated, where SNR=20 log(signal_metric / noise) dB. 7A to 10 illustrate the above calculations.

[0040] FIG. 7A shows a graph 700A of a moving average of a gradient waveform (e.g., graph 500B) representing normal aspiration, according to one or more embodiments, and FIG. 7B shows a graph 700B of a delta signal based on the moving average graph 700A, according to one or more embodiments. In some embodiments, the moving average is based on a moving average window of approximately 10 milliseconds (±10%). The delta signal graph 700B can include a noise floor window 702 from t=0 to 150 milliseconds, during which the delta signal can be used to calculate a noise floor amplitude. The delta signal graph 700B can also include a detection window 704, which in some embodiments has been determined to be optimal at 270 to 320 milliseconds. As described in more detail below, the delta signal within the detection window 704 is analyzed to determine whether the aspiration is normal or abnormal.

[0041] FIG. 8A shows a graph 800A of a moving average of a gradient waveform (e.g., graph 600B) representing an abnormal (low-sample) aspiration, according to one or more embodiments, and FIG. 8B shows a graph 800B of a delta signal based on moving average graph 800A, according to one or more embodiments. The moving average is again based on a moving average window of approximately 10 milliseconds (±10%), corresponding to the moving average window of normal aspiration in FIG. 7B. Delta signal graph 800B can also include a noise floor window 802 from t=0 to 150 milliseconds, within which the delta signal can be used to calculate a noise floor amplitude. Delta signal graph 800B can also include a detection window 804, corresponding to detection window 704, which has been determined to be optimal in some embodiments at 270 to 320 milliseconds. As described in more detail below, the delta signal within detection window 804 is analyzed to determine whether the aspiration is normal or abnormal.

[0042] Determining suitable detection windows and thresholds for determining normal and abnormal aspiration can be based on analysis of test samples of known normal and abnormal aspiration pressure signal waveforms.

[0043] 9 shows a graph 900 of multiple delta signals based on a moving average of eight sample aspiration pressure signal waveforms, where four are known to be normal aspirations and four are known to be abnormal aspirations (low-sample aspiration faults), according to one or more embodiments. The delta signals are based on a moving average window of approximately 10 milliseconds (±10%). The detection window 904 is optimally selected to be in the range of 270-320 milliseconds, as only abnormal aspirations will exhibit a delta signal signature there.

[0044] FIG. 10 shows a bar graph 1000 of SNR versus SNR metrics (average of absolute values, RMS, and 75th percentile) according to one or more embodiments for the delta signal of FIG. 9 , representing eight sample aspiration pressure signal waveforms S1, S2, S3, S4, S5, S6, S7, and S8. The calculated SNR (=20 log(signal_metric / noise) dB) for samples S1, S3, S5, and S7 represents a normal aspiration, while the calculated SNR for samples S2, S4, S6, and S8 represents an abnormal aspiration. Thus, an SNR threshold 1006 of 7 dB can be selected, with an SNR below 7 dB indicating a normal aspiration and an SNR above 7 dB indicating an abnormal aspiration (insufficient sample aspiration failure). The SNR threshold 1006 provides a clear boundary between a normal aspiration and an abnormal aspiration.

[0045] In some embodiments in which algorithm 320A is an AI algorithm, unsupervised learning methods such as K-means clustering can be used to identify abnormal suction in the pressure gradient waveform. The AI ​​algorithm 320A executable by processor 320P can be implemented in any suitable form of artificial intelligence programming, including, but not limited to, neural networks, including convolutional neural networks (CNNs), deep learning networks, regenerative networks, and other types of machine learning algorithms or models. It should be noted, therefore, that AI algorithm 320A is not, for example, a simple lookup table. Rather, AI algorithm 320A can be trained to detect or predict one or more types of suction disorders and can improve (make more accurate decisions or predictions) without being explicitly programmed.

[0046] FIG. 11 shows a graph 1100 of pressure gradient versus time for eight test sample aspiration pressure signal waveforms, according to one or more embodiments. Four test sample aspiration pressure signal waveforms represent normal aspirations, and four test sample aspiration pressure signal waveforms represent abnormal (under-sample) aspirations. Abnormal aspirations exhibit larger pressure gradients than normal aspirations (see boxed area 1108). Therefore, this “maximum slope” metric can be used in conjunction with K-means clustering to identify abnormal aspirations. That is, the maximum slope of the pressure gradient waveforms can be examined throughout the aspiration process to classify the aspiration as normal or abnormal (i.e., under-sampled). FIG. 12 shows a graph 1200 of a two-cluster classification of the maximum slope metric for eight test sample aspiration pressure signal waveforms (where the X-axis represents time in milliseconds and the Y-axis represents pressure gradient, which is the rate of change of normalized pressure over time), according to one or more embodiments. As shown, Cluster 1 (with center 1210) and Cluster 2 (with center 1212) are well separated, indicating that the maximum gradient metric is well suited for unsupervised (K-means clustering) classification of normal and abnormal (undersampled) aspirations.

[0047] Note that other unsupervised clustering methods may be used instead of K-means clustering, and supervised classification methods such as logistic regression, SVM (support vector machine), or Bayesian classifiers may also be used if the samples can be pre-labeled.

[0048] Returning to process block 404, method 400 can alternatively include analyzing the aspiration pressure measurement signal waveform via a processor executing an algorithm configured to derive a gradient waveform from the aspiration pressure measurement signal waveform by calculating the wavelet transform of the gradient waveform. As discussed above in connection with the moving average gradient waveforms of FIG. 7A (representing a normal aspiration) and FIG. 8A (representing an abnormal (under-sampled) aspiration), clear differences in the spectral signatures between the two waveforms are observable. In this analysis, the powerful simultaneous time scale (frequency) localization and multi-resolution analysis capabilities of wavelets are advantageously used. The wavelet transform may be a continuous wavelet transform (CWT) or a discrete wavelet transform (DWT).

[0049] An outline of the analysis using the CWT may include calculating the pressure gradient waveform from the aspiration pressure measurement signal waveform by calculating the difference of the pressure signals, as described above. A suitable moving average filter may be used to reduce noise amplification due to differentiation. The analysis may also include calculating the CWT of the pressure gradient signal in real time over a sliding time window as follows:

number

[0050] The analysis may further include examining the CWT coefficients at a particular range of scales, and then calculating a suitable metric based on the identified CWT coefficients and applying a suitable (specified) threshold to distinguish faulty aspiration from normal aspiration.

[0051] In some embodiments, calculating a suitable metric may include determining a baseline signal as follows: calculate the total CWT energy over a time window from 0 to t (e.g., chosen to be t=125 ms in this case) in a suitable scale range (determined to be, e.g., <13 in this case), then calculate the CWT energy within the same scale range (<13) for every time step of the detection window t>200 ms (determined as described below), or at sub-sampled time steps, and then calculate a detection SNR metric as follows:

number

[0052] FIG. 13 shows a graph 1300 of a CWT SNR metric versus time for a sample of normal aspiration, and FIG. 14 shows a graph 1400 of a CWT SNR metric versus time for a sample of abnormal aspiration, according to one or more embodiments. Based on the resulting CWT SNR metrics in graph 1300 (for the sample of normal aspiration) and graph 1400 (for the sample of abnormal aspiration), suitable detection windows 1314 and 1414 of t > 200 ms can be selected, and suitable SNR metric thresholds 1316 and 1416 of 7 dB can be selected. As shown by the region indicated by arrow 1417, which indicates SNR metric values ​​in detection window 1414 that exceed threshold 1416, these CWT SNR metrics are suitable for detecting abnormal (under-sample) aspiration. To maximize classification accuracy, additional sample data of normal and abnormal aspirations can be used to identify appropriate detection windows and CWT SNR thresholds.

[0053] In some embodiments of the CWT analysis, the following options can be considered:

[0054] The scale parameter 'a' and shift parameter 'b' can be restricted to discrete values, in particular to a binomial representation where the scale parameter is restricted to a power of two.

[0055] - Mallat's algorithm or Shensa's algorithm can be used with a binomial representation of the scale parameter. This can have a computational complexity of O(N), where N = the length of the signal vector. In practical detection, only signals with t > 200 ms may be considered, so N will be relatively small.

[0056] If a finer discretization of the scale parameter "a" is required (e.g., "a" = integer values ​​that do not need to be powers of 2), other methods may be used, such as spline-based fast CWT transform algorithms as in Unser et al., IEEE Trans on Signal Proc., 1994. These too can have a computational complexity of O(N).

[0057] By examining a larger dataset of normal and abnormal aspiration signal samples, the detection threshold, the preferred range of scale, the measure of the baseline signal, and the time window of the detection and baseline signals can be further adjusted.

[0058] Similarly, through further examination of more aspiration signal samples, the type of wavelet to be used can also be optimally selected.

[0059] One wavelet type that has been found suitable for use in this analysis may be the Symlet 2 wavelet, however, other suitable CWT types may also be used.

[0060] - Wavelet filters can be implemented using firmware or Data Manipulation Language (DML) level software and can be implemented on FPGA (Field Programmable Gate Array), DSP (Digital Signal Processor) chips, or other suitable ICs (Integrated Circuits).

[0061] In another embodiment, analysis of the aspiration pressure measurement signal waveform can include using a DWT. The advantage of using a DWT is its low computational cost and effectiveness in detecting transient phenomena (typically at lower scales) using multi-resolution analysis capabilities. A summary of the analysis using a DWT can include calculating a pressure gradient waveform from the aspiration pressure measurement signal waveform by calculating the difference of the pressure signals, as described above. A suitable moving average filter can be used to reduce noise amplification due to differentiation. The analysis can also include calculating the DWT of the pressure gradient signal in real time over a sliding time window, examining the DWT coefficients at a specific range of scales, calculating a suitable metric based on the identified DWT coefficients, and applying a suitable (specified) threshold to distinguish faulty aspiration from normal aspiration.

[0062] More specifically, the DWT analysis may involve determining a baseline signal as follows: Calculate the maximum DWT norm (largest coefficient value) over a time window from 0 to t (e.g., chosen to be t=125 ms in this case) in an appropriate scale range (determined to be, e.g., <2 in this case). Next, calculate the "DWT max norm" within the same scale range (<2) for every time step of the detection window t > 200 ms (determined as described below), or at subsampled time steps, and then calculate the detection SNR metric as follows:

number

[0063] FIG. 15 shows a graph 1500 of the DWT SNR metric versus time for a sample of normal aspiration, according to one or more embodiments, and FIG. 16 shows a graph 1600 of the DWT SNR metric versus time for a sample of abnormal aspiration, according to one or more embodiments. At lower scales (scale=1-2) over t>200 ms, a clear distinction in the DWT SNR metric is observed between the normal and abnormal aspiration samples. That is, the DWT SNR metric for the normal aspiration sample shown in FIG. 15 (and for several other normal aspiration samples similarly tested as described herein) was consistently below 0 dB. In contrast, the DWT SNR metric for the abnormal aspiration sample shown in FIG. 16 (and for several other abnormal aspiration samples similarly tested as described herein) was consistently at least 4 dB or greater for at least one time instance over t>200 ms. Based on these results, a suitable detection window 1514 and 1614 of t > 200 ms can be selected, and a suitable DWT SNR metric threshold 1516 and 1616 of 3 dB can be selected. (See, e.g., the area indicated by arrow 1617, which shows the DWT SNR metric values ​​of abnormal aspiration samples above threshold 1616 within detection window 1614.) To maximize detection accuracy, additional normal and abnormal aspiration sample data can be used to identify appropriate detection windows and DWT SNR thresholds.

[0064] FIG. 17 shows a schematic diagram of a multilevel DWT filter bank 1700 that can be used to implement the DWT analysis described above, according to one or more embodiments. The DWT filter bank 1700 can be a cascaded filter bank that includes separate high-pass and low-pass filtering and downsampling operations at each level. Here, "G" represents a high-pass filter, providing signal "detail" at each step at that scale, and "H" represents a low-pass filter. At each stage, the output of the H filtering is subsampled by half and passed through the next stage of wavelet filtering in the cascaded filter bank. In some embodiments, the DWT analysis described herein may require nine levels (note that only three are shown in FIG. 17). However, because transients in the pressure gradient signal (and consequently, abnormal suction) are detected using the lowest-scale DWT components, a complete analysis of the pressure gradient signal at every time step may not be necessary. Therefore, the DWT calculation may only need to be performed by the first level 1718 of the DWT filter bank 1700. This can advantageously result in significant computational runtime savings, thereby minimizing firmware or software overhead. In some embodiments, Symlet 2 or Daubechies order 2 and order 4 wavelets may be used. These can be implemented as simple second- or fourth-order low-pass and high-pass filters, as shown in FIG. 17.

[0065] Returning to FIG. 4 , method 400 can continue at process block 406 by identifying and responding to an aspiration fault (i.e., an under-sample aspiration fault) via the processor in response to the analysis performed at process block 404. As described above, an under-sample aspiration fault can be identified via spectral analysis by calculating a moving average, a conventional bandpass filter (e.g., a Butterworth filter), a CWT, or a DWT. Each of these analyses can accurately detect an under-sample aspiration fault by the end of the aspiration process. Method 400 can respond to an identified under-sample aspiration fault by timely terminating analysis of the liquid (involved in the under-sample aspiration fault) by the automated diagnostic analysis system prior to the start of any analysis of the liquid. In other embodiments, method 400 can alternatively or additionally respond to an identified under-sample aspiration fault by executing one or more other system procedures for an error condition.

[0066] Each of the three spectral analyses (moving average, CWT, and DWT) involves real-time calculations performed on a portion of the aspiration pressure signal waveform, thus advantageously limiting the size of the data stream analyzed at each time step. Each has a computational cost of O(N), making online implementation of these analyses feasible in firmware or software using a DSP (digital signal processor) microchip or FPGA (field programmable gate array).

[0067] While the disclosure is susceptible to various modifications and alternative forms, specific method and apparatus embodiments have been shown by way of example in the drawings and are herein described in detail. It is to be understood, however, that the specific methods and apparatus disclosed herein are not intended to limit the scope of the disclosure or the claims that follow.

Claims

1. 1. A method for detecting a sample shortage aspiration fault in an automated diagnostic analysis system, comprising: performing an aspiration pressure measurement via a pressure sensor in the automated diagnostic analysis system while the liquid is being aspirated; Derive the gradient waveform from the suction pressure measurement signal waveform: a moving average of the gradient waveform, or Wavelet transform of the gradient waveform analyzing the aspiration pressure measurement signal waveform via a processor executing an algorithm configured to calculate: In response to the analyzing, identifying and responding to the sample shortage aspiration fault via the processor; Including, The method, wherein the algorithm configured to calculate the moving average is further configured to calculate a difference between the moving average and the gradient waveform at predetermined time increments within a detection window of the gradient waveform.

2. 2. The method of claim 1, wherein the algorithm configured to calculate the moving average is further configured to calculate a signal-to-noise ratio equal to 20 log(signal_metric / noise) dB, wherein the signal_metric comprises at least one of an average, a root mean square (RMS), or a 75th percentile value of the absolute value of the difference between the moving average and the gradient waveform at predetermined time increments within the detection window of the gradient waveform.

3. 10. The method of claim 1, wherein the identifying and responding further comprises identifying and responding to the under-sample aspiration fault via the processor by determining whether a signal-to-noise ratio exceeds a threshold.

4. The method of claim 1 , wherein the moving average is based on a 10 millisecond moving average window.

5. The method of claim 1 , wherein the wavelet transform is a continuous wavelet transform (CWT) or a discrete wavelet transform (DWT).

6. The method of claim 1 , wherein the algorithm configured to compute the wavelet transform is further configured to compute a plurality of metrics of the wavelet transform based on wavelet transform coefficients.

7. 7. The method of claim 6, wherein the identifying and responding further comprises identifying and responding to the under-sample aspiration fault via the processor by determining whether a signal-to-noise ratio exceeds a threshold.

8. 2. The method of claim 1, wherein analyzing the aspiration pressure measurement signal waveform occurs during a detection window ranging from 270 milliseconds to 320 milliseconds from the start of the aspiration process.

9. 10. The method of claim 1, wherein the identifying and responding includes terminating analysis of the liquid by the automated diagnostic analyzer system in response to identifying the undersample aspiration fault before analysis of the liquid begins.

10. 1. An automated aspirating and dispensing device comprising: a robotic arm; a probe coupled to the robotic arm; a pump coupled to the probe; a pressure sensor configured to perform aspiration pressure measurements as liquid is being aspirated through the probe; a processor configured to execute an algorithm to analyze an aspiration pressure measurement signal waveform received from the pressure sensor by deriving a gradient waveform from the aspiration pressure measurement signal waveform and performing a spectral analysis of the gradient waveform by calculating a moving average or a wavelet transform of the gradient waveform; The automatic suction and dispensing device includes: The automated aspirating and dispensing device, wherein the algorithm is further configured to calculate a difference between the moving average and the gradient waveform at predetermined time increments within a detection window of the gradient waveform.

11. The automated aspirating and dispensing device of claim 10 , wherein the algorithm is further configured to calculate a plurality of metrics of the wavelet transform based on the wavelet transform coefficients.

12. 11. The automated aspirating and dispensing device of claim 10, wherein the algorithm is further configured to detect and respond to a sample shortage aspiration fault during the aspiration process by determining whether a signal-to-noise ratio exceeds a threshold.

13. The automatic aspirating and dispensing device according to claim 10, wherein the wavelet transform is a continuous wavelet transform (CWT) or a discrete wavelet transform (DWT).

14. the moving average is based on a 10 millisecond moving average window; or 11. The automated aspirating and dispensing device of claim 10, wherein the algorithm configured to perform the spectral analysis of the gradient waveform is further configured to perform the spectral analysis of the gradient waveform by calculating the moving average during a detection window ranging from 270 milliseconds to 320 milliseconds from the start of an aspiration process.

15. 11. The automated aspirating and dispensing device of claim 10, wherein the processor executing the algorithm is configured to respond to the sample shortage aspiration fault by terminating analysis of the liquid by the automated diagnostic analysis system during the aspiration process.

16. 1. An automated diagnostic analysis system comprising: The automatic suction and dispensing device according to claim 10; one or more analyzer stations for analyzing biological samples; an automated truck for transporting sample containers and reaction vessels to and from the automated aspirating and dispensing device and one or more analyzer stations; The automated diagnostic analysis system comprising:

17. 1. A non-transitory computer-readable storage medium comprising: a processor-executable algorithm configured to detect a sample-deficient aspiration fault based on a spectral analysis of a pressure gradient waveform derived from an aspiration pressure measurement signal waveform, the algorithm being configured to perform the spectral analysis of the pressure gradient waveform by calculating a moving average or a wavelet transform of the pressure gradient waveform; the algorithm is further configured to calculate a difference between the moving average and the gradient waveform at predetermined time increments within a detection window of the pressure gradient waveform.

Images