False Resistance Research Articles

Comparison of microscan MICroSTREP, PASCO, and sensititre MIC panels for determining antimicrobial susceptibilities of Streptococcus pneumoniae

The MicroScan MICroSTREP MIC panel was compared with PASCO and Sensititre systems against 157 isolates of Streptococcus pneumoniae chosen to include penicillin-susceptible, intermediate, and resistant strains. Arbitration testing was performed by microbroth dilution using National Committee for Clinical Laboratory Standards guidelines. Overall essential agreement of 94–97% and categorical agreement of 91–94% with the reference method was achieved for the three systems. There were 8 very major errors (false susceptibility) for PASCO, 10 for Sensititre, and 9 for MICroSTREP; 4 major errors (false resistance) each for PASCO and MICroSTREP, and 6 for Sensititre. Most of these errors occurred with trimethoprim/sulfamethoxazole. Minor errors (susceptible or resistant versus intermediate) totaled 47 for PASCO, 69 for Sensititre, and 53 for MICroSTREP. Minor interpretive errors were most common with penicillin and ceftriaxone. This study showed that all three MIC panels provided interpretive results comparable to one another and to the reference method.

Quality of antimicrobial susceptibility testing in the UK: a Pseudomonas aeruginosa survey revisited.

As part of a programme to assess the usefulness of routine antimicrobial susceptibility data as a surveillance tool, we reviewed the results of a national survey of resistance in Pseudomonas aeruginosa, undertaken in 1993. Twenty-four UK laboratories contributed isolates for centralized MIC testing, indicating also their own susceptibility test data. As reported previously (Chen et al. (1995) Journal of Antimicrobial Chemotherapy 35, 521-34), the rate of false resistance (isolates reported susceptible, but found resistant on MIC testing/all isolates reported susceptible) was 0.6-8%, according to the antimicrobial and breakpoint. Review showed that this favourable position reflected the fact that >88% of isolates were susceptible to any given antimicrobial and--in most cases--were correctly reported as such. Reporting was more erratic for resistant isolates: for beta-lactams and amikacin, isolates resistant at the highest MIC breakpoints were equally likely to be reported as 'susceptible' or 'resistant'; such misreporting was less common with ciprofloxacin and gentamicin but still occurred in 9-20% of cases. Conversely, up to 73% of the isolates reported as resistant proved to be susceptible at high breakpoints, and up to 44% were susceptible at low breakpoints. Miscategorizations did not reflect failure to detect particular mechanisms but, rather, the fact that MIC and zone breakpoints for P. aeruginosa serve to cut 'tails' of resistant organisms from continuous distributions, not to distinguish discrete populations. In this situation, some disagreement between routine tests and MICs is inevitable, but the frequency at which highly resistant isolates were reported as sensitive is disturbing. For surveillance, we conclude that resistance rates based on routine tests are unreliable for P. aeruginosa. This situation may improve with greater standardization of routine testing, but the continuous susceptibility distributions without discrete resistant and susceptible populations militate against perfect agreement. Despite these deficiencies, routine data should allow trend analysis.

Evaluation of the Vitek System to accurately test the susceptibility of Pseudomonas aeruginosa clinical isolates against cefepime

A total of 300 recent blood stream and pneumonia isolates of Pseudomonas aeruginosa from 32 different medical centers in the United States were used to assess the accuracy of the Vitek System (GNS-107 card) for cefepime, a new “fourth-generation” cephalosporin. The Vitek System test result was compared to the consensus of the three other methods (reference broth microdilution, disk diffusion, Etest (AB BIODISK, Solna, Sweden)), and 271 of 300 consensus category of susceptibility sets were unanimous. The Vitek System produced a 25.3% error rate (5.3% false resistance, 20.0% minor errors). A consistent trend toward higher MIC results with the Vitek System was observed that produced a 15.3 to 21.3% lower susceptible rate compared with the other susceptibility test methods. The consensus cefepime resistance rate was only 4.3% compared to 14.0% for the Vitek GNS-107 card. The error was reproducible on triplicate repeat testing. These results indicate an unacceptable rate of false resistance being produced by the Vitek System when testing cefepime against P. aeruginosa strains. Alternative methods are suggested for this broad-spectrum antipseudomonal cephalosporin.

Direct susceptibility testing with positive BacT/Alert blood cultures by using MicroScan overnight and rapid panels.

Studies were conducted on a method of direct inoculation of MicroScan dried overnight and of rapid panels with positive aerobic blood cultures obtained from the BacT/Alert to determine antimicrobial susceptibilities. Inocula were limited to specimens that appeared unimicrobic on Gram stain. Results were compared to those obtained from panels inoculated following subculture. For 133 gram-negative bacilli, there were 94.7 and 93.5% categorical agreements between direct and standard methods for all drugs tested with overnight and rapid panels, respectively. For 104 gram-positive cocci, there were 93.2 and 93.1% categorical agreements for overnight and rapid panels, respectively. The major error (false resistance) rate for gram negatives was 1.4% for overnight versus 0.7% for rapid panels. The very major error (false susceptibility) rate was 2.7% for overnight versus 8.1% for rapid panels. The total error rates were 1.6% for overnight panels and 1.5% for rapid panels. The major error rates for gram-positive direct susceptibility tests were 2.6% for overnight and 2.5% for rapid panels. The very major error rates were 8.8 and 7.2% for overnight and rapid panels, respectively. Total error rates were 3.6% for overnight and rapid gram-positive panels. These findings suggest that susceptibility results obtained from directly inoculated gram-negative overnight panels have the greatest correlation to those obtained by standard methods. When discrepant results occur with direct-susceptibility testing, they are more likely to show false susceptibility than false resistance.

Novel antibiotic susceptibility tests by the ATP-bioluminescence method using filamentous cell treatment.

Antimicrobial susceptibility testing by the ATP-bioluminescence method has been noted for its speed; it provides susceptibility results within 2 to 5 h. However, several disagreements between the ATP method and standard methodology have been reported. The present paper describes a novel ATP method in a 3.5-h test which overcomes these deficiencies through the elimination of false-resistance discrepancies in tests on gram-negative bacteria with beta-lactam agents. In our test model using Pseudomonas aeruginosa and piperacillin, it was shown that ATP in filamentous cells accounted for the false resistance. We found that 0.5% 2-amino-2-methyl-1,3-propanediol (AMPD) extracted ATP from the filamentous cells without affecting normal cells and that 0.3 U of adenosine phosphate deaminase (APDase)/ml simultaneously digested the extracted ATP. We used the mixture of these reagents for the pretreatment of cells in a procedure we named filamentous cell treatment, prior to ATP measurements. This novel ATP method with the filamentous cell treatment eliminated false-resistance discrepancies in tests on P. aeruginosa with beta-lactam agents, including piperacillin, cefoperazone, aztreonam, imipenem-cilastatin, ceftazidime, and cefsulodin. Furthermore, this novel methodology produced results which agreed with those of the standard microdilution method in other tests on gram-negative and gram-positive bacteria, including P. aeruginosa, Escherichia coli, Staphylococcus aureus, and Enterococcus faecalis, for non-beta-lactam agents, such as fosfomycin, ofloxacin, minocycline, and aminoglycosides. MICs obtained by the novel ATP method were also in agreement with those obtained by the agar dilution method of susceptibility testing. From these results, it was shown that the novel ATP method could be used successfully to test the activities of antimicrobial agents with the elimination of the previously reported discrepancies.

Multicenter laboratory evaluation of the bioMérieux Vitek antimicrobial susceptibility testing system with 11 antimicrobial agents versus members of the family Enterobacteriaceae and Pseudomonas aeruginosa.

A four-center study in which a total of 1,082 recent clinical isolates of members of the family Enterobacteriaceae and Pseudomonas aeruginosa were examined versus 11 antimicrobial agents with the bioMérieux Vitek susceptibility test system (Hazelwood, Mo.) and the GNS-F6 card was conducted. In addition, a challenge set consisting of the same 200 organisms was examined in each of the four participating laboratories. Results obtained with the Vitek system were compared to MICs determined by a standardized broth microdilution method. For purposes of comparison, susceptibility categories (susceptible, intermediate, or resistant) were assigned on the basis of the results of both methods. The result of the broth microdilution test was considered definitive. The total category error rate with the Vitek system and the recent clinical isolates (11,902 organism-antimicrobial comparisons) was 4.5%, i.e., 1.7% very major errors, 0.9% major errors, and 1.9% minor errors. The total category error rate calculated from tests performed with the challenge set (i.e., 8,800 organism-antimicrobial comparisons) was 5.9%, i.e., 2.2% very major errors, 1.1% major errors, and 2.6% minor errors. Very major error rates higher than the totals were noted with Enterobacter cloacae versus ampicillin-sulbactam, aztreonam, ticarcillin, and ticarcillin-clavulanate and with P. aeruginosa versus mezlocillin, ticarcillin, and ticarcillin-clavulanate. Major error rates higher than the averages were observed with Proteus mirabilis versus imipenem and with Klebsiella pneumoniae versus ofloxacin. Excellent overall interlaboratory reproducibility was observed with the Vitek system. The importance of inoculum size as a primary determinant in the accuracy of susceptibility test results with the Vitek system was clearly demonstrated in this study. Specifically, when an inoculum density fourfold higher than that recommended by the manufacturer was used, high rates of false resistance results were obtained with cell wall-active antimicrobial agents versus both the Enterobacteriaceae and P. aeruginosa.

Critical evaluation of the Vitek GNS F6 card results compared to standardized, reference susceptibility test methods

A large number of Enterobacteriaceae (291 unselected and 30 clinical challenge isolates) were used to evaluate the accuracy of the Vitek GNS-F6 susceptibility testing card for 10 antimicrobial agents (ampicillin, ampicillin/sulbactam, aztreonam, ciprofloxacin, imipenem, mezlocillin, ofloxacin, piperacillin, ticarcillin, and ticarcillin/clavulanate). Results were compared to reference broth microdilution and disk diffusion methods. Highest interpretive error on initial processing were observed with ticarcillin/clavulanate (very major false-susceptible error, 3.4%), imipenem (major false-resistant error, 1.9%), and ampicillin/sulbactam (minor error, 15.9%). Repeat testing resolved many very major errors (21 of 41 results), and the overall accuracy rate was improved from 92.1 to 93.4%. Analysis of all minor interpretive errors demonstrated that the trend for Vitek was to report slightly lower MICs (6 of 10 drugs; 97 of 159 results) in comparison to the reference test results. If a heavy inoculum was used in the Vitek cards, false resistance was observed more frequently with aztreonam and penicillins. These data demonstrate that susceptibility testing accuracy with Vitek GNS-F6 card based on categorical agreement varies from 83.2% (ampicillin/sulbactam) to 99.4% (ciprofloxacin) when testing enteric bacilli. Some antimicrobial agents (beta-lactamase inhibitor/penicillin combinations, antipseudomonal penicillins) may require slight modification of interpretive software. Finally, the overall accuracy of the Vitek System was greater than 91% for 9 of 10 drugs tested with a very low, acceptable rate of false-susceptible error (0.8%).

The reliability of Microscan™ conventional and rapid panels to identify Staphylococcus aureus and detect methicillin resistance: An evaluation using the tube coagulase test and mecA PCR

Microscan (Dade Diagnostics, Brisbane) Positive Combo Type 6 (312 panels) and Rapid Positive Breakpoint Type 1 (62 panels) were evaluated for Staphylococcus aureus identification, using the tube coagulase test (TC), and oxacillin susceptibility, using mecA. A total of 374 consecutive clinical staphylococci were tested, with TC and Microscan having 100% correlation (335 identified as S. aureus and 39 as coagulase negative staphylococci by both methods). A 93% correlation was observed between Microscan and mecA PCR for oxacillin susceptibility. No very major errors (0/374 false oxacillin susceptibility) and 26 (7%) major errors (26/374 false oxacillin resistance) were found showing false resistance to oxacillin to be a problem in our population. Oxacillin Etest (AB Biodisk, Sweden) was performed on all oxacillin resistant isolates. A bimodal distribution was observed between mecA positive and negative isolates. A testing algorithm (using the Microscan panels and Etest) was developed for this laboratory to detect mecA encoded methicillin resistance. Retrospective application of this algorithm to the 374 isolates gave 100% correlation with mecA detection.

Accurate characterization of ofloxacin susceptibility with Enterobacteriaceae using a modified GNS F6 card and the bioMerieux Vitek System

Routine antimicrobial susceptibility testing of Enterobacteriaceae using the Vitek System (bioMerieux Vitek, Hazelwood, MO) and the GNS-F6 card revealed discrepancies between the activity of ofloxacin and ciprofloxacin, primarily with isolates of Klebsiella pneumoniae. Specifically, during a one-year period, 12% of 618 ciprofloxacin-susceptible isolates were determined to be ofloxacin resistant with the GNS-F6 card. A similar problem, but one of lower magnitude, was observed with Serratia marsescens. That these represented false ofloxacin resistance results was confirmed by comparison of broth microdilution determinations of ofloxacin MICs with F6 results on a collection of 203 fresh clinical isolates of K. pneumoniae and 39 isolates of S. marsescens. The GNS-F6 card was then modified by the manufacturer to include a new formulation of ofloxacin and assessed using a collection of 224 recent clinical isolates of Enterobacteriaceae and Pseudomonas aeruginosa, and 78 stock cultures of enteric Gram-negative bacilli selected specifically because of disproportionately high rates of fluoroquinolone resistance. No ofloxacin false resistant results were observed with this collection when the modified GNS-F6 card was evaluated in comparison to a standardized broth microdilution MIC test. The current clinical version of the Vitek System appears to accurately assess ofloxacin susceptibility.

Detection of oxacillin-resistance in Staphylococcus aureus by MicroScan MIC panels in comparison to four other methods

Two hundred fifty-two isolates of Staphylococcus aureus were tested for oxacillin susceptibility by MicroScan Gram positive overnight and rapid MIC panels. Results were compared with nonautomated methods including disk diffusion, MRSA Crystal ID, and Etests using MRSA Screen Agar as reference. One hundred sixty-nine isolates (67.1%) were oxacillin-susceptible and 83 (32.9%) were resistant. All methods agreed for 234 (92.9%) isolates. Very major error rates were 1.2% for disk diffusion, 3.6% for Etest, and 0 for all other methods. Major error rates were 5.3% for MicroScan overnight panels, 3% for rapid panels, 2.4% for disk diffusion, 1.2% for Etest, and 0.6% for MRSA Crystal ID. Nine oxacillin-susceptible isolates with borderline MICs and discrepant results for 1 or more methods were tested for the mec A gene and all were negative. Each was susceptible to β lactam/β lactamase inhibitor combinations, suggesting that false resistance may have been due to excessive β lactamase production. Oxacillin-resistant S. aureus with borderline MICs determined by MicroScan should be confirmed by an alternate method. The most practical and cost-effective means among those we tested is the MRSA Screen Agar.

Testing of susceptibility of Mycobacterium tuberculosis to pyrazinamide: comparison of Bactec method with pyrazinamidase assay

The susceptibility of 428 clinical isolates of Mycobacterium tuberculosis to pyrazinamide was assessed by the Bactec method and the Wayne pyrazinamidase assay. The correlation between the two tests was 98.2 and 100% for susceptible and resistant strains, respectively. False resistance was seen in four (0.8%) strains with the Bactec test, and false-susceptible results occurred in two (0.5%) pyrazinamidase assays. The Bactec test is rapid and reliable, and the Bactec results correlate well with the pyrazinamidase test results, although some strains did not grow well in the test medium.

Investigation of the reformulated Remel Synergy Quad plate for detection of high-level aminoglycoside and vancomycin resistance among enterococci.

We investigated the accuracy of the recently released Remel Synergy Quad plate, a commercially available agar screening method for detecting high-level aminoglycoside and vancomycin resistance among enterococci that is based on the National Committee for Clinical Laboratory Standards recommended guidelines (National Committee for Clinical Laboratory Standards, M7-A3, 1993). The Synergy Quad correctly determined the gentamicin and streptomycin resistance status for > or = 97% of 147 Enterococcus faecalis and Enterococcus faecium isolates tested. Detection of vancomycin resistance also was reliable, as no false susceptibility occurred with 36 vancomycin-resistant E. faecalis and E. faecium strains and false resistance occurred only once with the 47 susceptible strains tested. One strain each of Enterococcus gallinarum and Enterococcus casseliflavus failed to grow on the screen, but because the true nature and significance of resistance in such isolates is unknown the implication of their screen negativity is uncertain. In summary, the Remel Synergy Quad provides a highly accurate and convenient method for susceptibility testing of enterococci against gentamicin, streptomycin, and vancomycin.

Prevalence of enterococcal high-level aminoglycoside resistance in Japan comparative detection by three methods

A total of 250 strains of enterococci isolated in Kumamoto University Hospital, Japan, during the period from January to March 1992 were tested for high-level aminoglycoside resistance. Brain-heart infusion (BHI) agar plates supplemented with 1000 μg/ml of gentamicin or 2000 μg/ml of streptomycin detected 164 (66%) isolates resistant to either gentamicin or streptomycin alone, or both, and consisted of 107 (43%) resistant to gentamicin and 96 (38%) resistant to streptomycin. The Vitek Gram-Positive Susceptibility card (GPS-TA) revealed high correlations with those by agar screens, the results indicating a sensitivity of 100% and 99% to gentamicin and streptomycin, respectively, and 100% specificity to both. Also, the microdilution tests of the National Committee for Clinical Laboratory Standards (NCCLS) showed 100% and 92% sensitivity to gentamicin and streptomycin, respectively, and no false resistance (100% specificity) when compared with the results by agar screens.

Interpretive criteria for CI-960 (AM-1091, PD127391) disk diffusion tests using 5-μg disks

CI-960, a new potent fluoroquinolone, was evaluated to develop disk diffusion susceptibility criteria for a 5-μg content disk. Preliminary interpretive zone diameter criteria were proposed for two possible susceptible breakpoint MICs (⩽1 and ⩽2 μg/ml). Both sets of interpretive criteria performed well having 96.8%-98.4% absolute agreement and no false-susceptible errors. The more conservative ⩽1- μg/ml breakpoint criteria had correlate zones of ⩾23 mm for susceptible and ⩽19 mm for resistant. These criteria produced no false resistance, whereas 0.7% of strains exhibited major errors at the ⩽2- μg/ml breakpoint.

Detection of high-level aminoglycoside resistance in enterococci other than Enterococcus faecalis

The ability of six screening methods to detect high-level aminoglycoside resistance in enterococcal species other than Enterococcus faecalis was investigated. The 85 Enterococcus isolates, which included 55 E. faecium, 11 E. gallinarum, 9 E. casseliflavus, 5 E. raffinosus, 4 E. avium, and 1 E. mundtii, were tested by using aminoglycoside-supplemented brain heart infusion agar (BHI), Remel EF Synergy Quad plates, high-content aminoglycoside diffusion disks, standard (prepared in-house) microdilution panels, Pasco MIC Gram Positive microdilution panels, and Vitek GPS-TA cards. When tested on BHI, 32 and 35 strains showed resistance to gentamicin and streptomycin, respectively. Resistance profiles obtained with Remel EF Synergy Quad plates were in complete agreement with those obtained on BHI. However, growth on Mueller-Hinton agar-based plates was not as heavy. Some isolates showed only weak growth and required 48 h for resistance to become evident, especially with swab inoculation of quadrants containing 2,000 micrograms of gentamicin per ml. Profiles obtained by use of the agar-based screens were used as the basis for evaluating the other methods. Disk diffusion showed complete agreement. No false resistance occurred by either microdilution method, but 48 h of incubation was needed for detection of some gentamicin-resistant isolates, and 14% of the streptomycin-resistant strains were not detected by standard microdilution. The Vitek GPS-TA card detected 81 and 100% of the gentamicin- and streptomycin-resistant isolates, respectively. In general, most methods used to detect high-level aminoglycoside resistance in E. faecalis appear to be reliable for the testing of the other enterococcal species. However, further investigations with a greater number of resistant E. raffinosus, E. avium, and E. mundtii isolates, when they are available, will be useful for establishing the full range of enterococci that can reliably be tested by the various methods.

Factors influencing determination of high-level aminoglycoside resistance in Enterococcus faecalis

The ability of seven methods to detect high-level gentamicin (58 strains) and streptomycin resistance (56 strains) among 107 Enterococcus faecalis isolates was investigated at the University of Chicago Medical Center and the University of Nebraska Medical Center. Methods included a standard agar screen plate, high-content disk diffusion, Remel (Lenexa, Kans.) EF Synergy Quad plates, standard microdilution panels prepared in house, Pasco MIC Gram-Positive panels (Difco Laboratories, Detroit, Mich.), MicroScan MIC Type 5 dry panels (Baxter Healthcare Corp., MicroScan Div., West Sacramento, Calif.), and Vitek GPS-TA cards (Vitek Systems Inc., Hazelwood, Mo.). Results indicating false resistance were not obtained by any method, and there was 100% agreement between the results of the disk diffusion and standard agar screen methods. Prolonging incubation from 24 to 48 h increased resistance detection for both agar and microdilution screens. EF Synergy Quad plates inoculated with micropipettes detected 100% of the streptomycin- and gentamicin-resistant isolates. Resistance detection for streptomycin and gentamicin, respectively, was 93 and 96% by standard microdilution, 93 and 98% by Pasco panels, 88 and 89% by MicroScan panels, and 88 and 91% by Vitek GPS-TA cards. False susceptibility occurred more frequently with streptomycin-resistant isolates than it did with gentamicin-resistant strains and appeared to be strain related in some instances. The use of an increased inoculum size enhanced resistance detection with these strains, but it complicated interpretation of results and led to the selection of streptomycin-resistant mutants. Until results of further studies delineate optimum test conditions, a delay in the final interpretation of agar and microdilution screen results until 48 h for isolates showing no or light growth at 24 h may help to minimize the occurrence of false susceptibility reporting.

Comparison of 30 μg and 120 μg gentamicin disks for the prediction of gentamicin resistance in Enterococcus faecalis

Single strain regression analysis was performed on PDM II medium for E. faecalis with 10, 30 and 120 micrograms gentamicin disks using E. faecalis, strain ATCC 29212 as the reference. This method permits the calculation of zone diameters corresponding to different MIC values for different disk contents. The lack of discrimination between normal low-level resistant strains and high-level resistance using the 10 micrograms disk was confirmed. However, both the 30 micrograms and 120 micrograms disks seemed to provide a separation of the normal low-level gentamicin-resistant population from strains with increased resistance. Since the 30 micrograms disk is used routinely in some countries, there should be no need for an extra high content disk in these laboratories. This was confirmed when 96 clinical isolates of E. faecalis were analysed and the results of routine disk diffusion tests were compared with the MIC values. Two of the strains showed high-level gentamicin resistance (greater than 2000 mg/l) and produced no zone of inhibition. The other 94 isolates showed gentamicin MIC values between 4-16 mg/l, and 72 of the MIC results were 8 mg/l. The zone diameters for these strains ranged between 15 and 25 mm with a mean of 18.2 and a median value of 18 mm. In order to include statistical considerations of the zone size populations for setting of breakpoints, a study of gentamicin zone size distributions was performed for several bacterial species. Inhibition zone diameter values around the 30 micrograms gentamicin disk for 2079 clinical isolates of E. faecalis, 2268 S. aureus, 3201 E. coli and 547 strains of P. mirabilis from different years were plotted as histograms. Tests for agreement with a Gaussian distribution showed that the histograms were slightly peaked and skewed towards higher zone values. Parametric and non-parametric statistical tests were compared and the results showed that means and medians were very similar and that parametric fractile estimations at the lower end of the histogram populations were conservative and could be used in view of a slightly lower rate of false resistance. The 1% parametric fractile of 12 mm was selected as a suitable breakpoint for the identification of normal, low-level resistant isolates of E. faecalis using the standardized disk test of the Swedish Reference Group for Antibiotics.

False resistance to imipenem with a microdilution susceptibility testing system

Routine monitoring of antibiotic resistance at Children's Hospital, Boston, detected a dramatic increase in the prevalence of imipenem-resistant strains of Pseudomonas aeruginosa. Further studies documented that false resistance to imipenem was due, in part, to the loss of imipenem potency in customized MIC microdilution trays supplied by Sensititre Ltd. (West Sussex, United Kingdom). Recognition of the problem was delayed by use of the quality control standard recommended by the manufacturer, which were higher and broader than those suggested by the National Committee for Clinical Laboratory Standards.

A Fire Detection Algorithm Using Second Order Statistics

A detection algorithm is presented, based on the comparison of the short- and long-term power spectral densities of monitored input signal sequences from Tire sensors'. The detector adapts slowly varying changes in the f i t and second order statistics of the monitored signals, which may occur due to environmental influences. The algorithm is as well suited for single- and multiple input detection. A recursive implementation reduces the calculation effort. Experimental results are presented, concerning the detection capability and the false alarm resistance of the detection algorithm in realistic environmental conditions.

A composite detection algorithm using signal trend information of two different sensors

A detection algorithm, based on the well known two-input Kendall-Tau trend detector is presented. The necessary modifications and preprocessing requirements for fire detection purposes are discussed. Experimental results are shown, concerning the detection capability and the false alarm resistance of the detection algorithm under realistic environmental conditions. A method for automated optimization of decision thresholds in a given installation environment is proposed.