Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Accuracy”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 757 records · Page 42Linked to original sources
PubMed1999

Accuracy of endorectal ultrasound after preoperative radiochemotherapy in locally advanced rectal cancer.

OBJECTIVES: Factors limiting the accuracy of endorectal ultrasound in staging, locally advanced primary rectal cancer after preoperative neoadjuvant radiochemotherapy (RCT) were evaluated. METHODS: Patients (n = 84) with initial locally advanced rectal cancer (uT3/uT4) undergoing R0 resection were investigated after preoperative treatment that combined radiotherapy up to 45 Gy with two cycles of chemotherapy (5-FU and leucovorin on d 1-5 and 22-28). At 4 to 6 weeks after completion of RCT and before tumor resection, preoperative endoluminal ultrasound was performed. RESULTS: The accuracy to predict the depth of tumor infiltration (T-category) was found to correlate with downstaging. The T-category was correctly staged before surgery in 15 of the 51 responders (29%) and in 27 of 33 nonresponders (82%), whereas misinterpretation occurred in 36 of the responders (71%) and in 6 of the nonresponders (18%) (p < 0.001). Neither tumor distance from anal verge nor tumor location correlated with the staging accuracy. Lymph node involvement was correctly assessed in 48 patients (57%). Wall invasion was correctly ascertained in 42 patients (50%), with under estimation in 11 patients (13%) and overestimation in 31 patients (37%). CONCLUSIONS: After radiochemotherapy, endosonography does not provide a satisfactory accuracy for preoperative staging of rectal cancer. New interpretation and diagnostic criteria are needed for the prediction of treatment response.

Adenocarcinoma↗
PubMed2004

Accuracy of preoperative pinhole subtraction single photon emission computed tomography for patients with primary and recurrent hyperparathyroidism in an endemic goiter area.

PURPOSE: Bilateral parathyroid exploration is still the standard therapeutic procedure for primary and recurrent hyperparathyroidism (HPTH). Since a unilateral surgical strategy that reduces surgical complications should not increase the risk of missing enlarged parathyroid glands, reliable preoperative imaging is the first requirement for this approach. This study was conducted to assess the accuracy of preoperative (99m)TcO(4)-(201)T1 pinhole subtraction single photon emission computed tomography (SPECT) compared with sonography. METHODS: The study population consisted of 15 patients with primary (n = 13) or recurrent (n = 2) HPTH who underwent preoperative (99m)TcO(4)-(201)T1 pinhole subtraction SPECT. Preoperative sonography was also done in 14 of these patients. RESULTS: (99m)TcO(4)-(201)T1 pinhole subtraction SPECT was significantly more accurate than sonography. It had an overall accuracy of 80% in detection of enlarged parathyroid glands. The accuracy of localization increased to 92.3% in patients with solitary adenomas, and to 100% in those with solitary adenomas and no previous parathyroid exploration. CONCLUSIONS: The high accuracy of preoperative localization with (99m)TcO(4)-(201)T1 pinhole subtraction SPECT in patients with primary and recurrent HPTH allows for an imaging-guided unilateral operative strategy in most patients, even those with concomitant nodular goiters. This may reduce the risk of surgical complications and expand the use of minimally invasive techniques in parathyroid surgery.

Adenoma↗
PubMed2004

Real-time integration of ultrasound into neuronavigation: technical accuracy using a light-emitting-diode-based navigation system.

BACKGROUND: In brain surgery, intraoperative brain deformation is the major source of postimaging inaccuracy of neuronavigation. For intraoperative imaging of brain deformation, we developed a platform for the integration of ultrasound imaging into a navigation system. METHOD: A commercially available ultrasound system was linked to a light-emitting-diode- (LED) based neuronavigation system via rigid fixation of a position localiser to the ultrasound probe and ultrasound image transfer into the navigation system via a S-VHS port. Since the position of the ultrasound image co-ordinate system is not readily defined within the navigation reference co-ordinate system (REF CS), a transformation which links both co-ordinate systems has to be defined by a calibration procedure. Calibration of the ultrasound probe within the REF CS was performed via a cross-wire phantom. The phantom target was defined within the navigation co-ordinate system (by pointer under microscopic control) and imaged by ultrasound. Ultrasound presets were optimised (digital beam focusing, gain intensity) to attain a small echoic target for manual target definition. The transformation was derived from 150 ultrasound measures and iteration. Accuracy was calculated as mean linear error (LE; in X(REF), Y(REF), or Z(REF) direction), overall mean LE (linear errors of all axes X(REF) to Z(REF)) and Euclidean error (EE; vectorial distance from the physical target). FINDINGS: Optimised ultrasound presets (8 MHz frequency, digital beam focusing, 20% gain intensity) enabled a low interobserver error (mean: 0.5 mm, SD: 0.28) for target definition within the 2-D ultrasound image. Mean accuracy of pointer-based physical target definition in the REF CS was 0.7 mm (RMSE; SD: 0.23 mm). For navigated ultrasound, the overall mean LE was 0.43 mm (SD: 1.36 mm; 95%CL: 3.13 mm) with a mean EE of 2.26 mm (SD: 0.97 mm; 95%CL: 4.21 mm). INTERPRETATION: Using a single target cross-wire phantom, a highly accurate integration of ultrasound imaging into neuronavigation was achieved. The phantom accuracy of integration lies within the range of application accuracy of navigation systems and warrants clinical studies.

Brain↗
PubMed1998

[Stereolithography models vs. milled 3D models. Production, indications, accuracy].

Anatomic, life-like, three-dimensional models have a definite place in cranio-maxillofacial surgery. Our experience with 541 computer tomographic (CT)-based three-dimensional (3D)-models for diagnostic purposes, preoperative planning and model operations in our department was gained using stereolithographic and milled models. The question of which production method is preferable is a matter of controversy in the literature. Both methods are based on CT and magnetic resonance imaging data, but the differences in production give rise to specific advantages and disadvantages. For comparison we scanned two measurement models, fabricated milled and stereolithographic models, and analyzed the differences concerning accuracy and shape of specific structures. The scan distance (feed) is the limiting factor for the accuracy for both methods. Milled models show the highest precision in the plane of CT scanning--the more oblique the measurements became to this plane, the greater the decrease in accuracy. Hollows and undercuts can only be produced by splitting the model. Stereolithographic models show higher deviations with randomly distributed errors. The mean deviation was 0.81 mm on stereolithography and 0.54 mm on milled models. The accuracy of both methods is sufficient for clinical use. In routine cases the milling method seems to be superior because of shorter production time and lower costs. In special cases, where hollows and fine structures play a major role, stereolithography is the method of choice.

Humans↗
PubMed2004

Accuracy of endorectal ultrasonography in the preoperative staging of rectal cancer.

BACKGROUND: Preoperative staging of rectal tumours is considered essential to tailoring treatment for individual patients. The aim of the present study was to evaluate the accuracy of 3-D-endorectal ultrasonography for the preoperative staging of rectal cancer. METHODS: Three hundred and fifty-seven patients with rectal adenocarcinomas underwent an endorectal ultrasonography evaluation during a period of eight years. The evaluation was performed by four surgeons. We compared the endorectal ultrasonography staging with the pathology findings. Patients with preoperative chemoradiation were excluded from the study. RESULTS: Overall accuracy in assessing the level of rectal wall invasion was 77.3%, with 9.3% of the tumours overstaged and 8.1% understaged. Accuracy in assessing nodal involvement in 313 patients treated with radical surgery was 74.9%, with 8.9% overstaged and 8.9% understaged. CONCLUSIONS: The accuracy of 3-D-endorectal ultrasonography in assessing the depth of tumour infiltration is good, but it is lower than previously reported. The technique is precise in distinguishing between benign tumours and invasive cancer. The results depend on the experience of the surgeon.

Adenocarcinoma↗
PubMed2002

Interpretation accuracy of a CCD film digitizer.

There are no significant differences between the interpretations of radiographic images resulting from digitizing films using a recently developed CCD unit and the readings of the original films as measured by accuracy, sensitivity, specificity and ROC analysis. Digital imaging is rapidly becoming the basis of radiology practice resulting in the gradual elimination of conventional film examinations. Related to this trend is the ability of radiologists to provide reliable interpretations of imaging examinations using film digitization and soft copy display. An acceptable system must provide high-quality digital images to achieve the levels of diagnostic accuracy that are comparable to the interpretation of film. This study is an effort to contribute to the acceptance of digital imaging by testing the hypothesis.- The authors selected 120 radiographic examinations from the departmental film files which included chest, abdominal, extremity and other common procedures. Half of the cases contained specified abnormalities and half did not have such findings. All of the examinations had a high degree of diagnostic difficulty. The films were digitized on the CCD equipment and each of the four board certified radiologists who participated in the study interpreted each of the 120 cases, half on original film and half in the digitized format on a high resolution workstation. No radiologist read the same examination more than once. Data collection included variables such as perceived image quality, diagnostic difficulty and interpretation confidence. Accuracy measures were calculated and an ROC analysis was performed. As of this date, the preliminary results indicate acceptance of the hypothesis that there is no significant difference between the accuracy of film readings and the interpretations of the digitized images on a soft copy display. Differences noted in perceived image quality among readers were also not significant. The final results, including an ROC analysis, will be available in March 2002 when a paper is submitted for publication in the Journal of Digital Imaging. During the transition from conventional film practice to digital radiology there is a need to convert previous films to a digital format for comparison with new digital examinations and to facilitate expert reading and consultation on film procedures performed at distant sites. Diagnostic quality and cost effectiveness are key factors in the acceptability of film digitizers to meet these requirements. The preliminary results of this study are encouraging in that another option may be available to radiologists and administrators responsible for choosing equipment for the changing pratice of medical imaging.

Data Display↗
PubMed2006

Accuracy and underestimation of malignancy of breast core needle biopsy: the Florence experience of over 4000 consecutive biopsies.

Breast core needle biopsy (CNB) is used for sampling breast lesions in both the screening and diagnostic context. We present the accuracy of breast CNB from a consecutive series of 4035 core biopsies, using methods that minimise selection and verification bias. We calculate accuracy and underestimation of malignancy for both automated (14G) and directional vacuum-assisted (11G) CNB performed under stereotactic or sonographic guidance. Overall sensitivity of CNB is 94.2% (92.9-95.5%) and specificity is 88.1% (86.6-89.6%), positive and negative predictive values are 84.8% (82.9-86.7%) and 95.6% (94.6-96.6%), respectively. In sampling microcalcification, the overall underestimation of malignancy is 26.6% (22.9-30.3%): underestimation is significantly higher for automated CB relative to VAB (chi2 ((df = 1)) = 8.90 , P = 0.002), the absolute difference in underestimation being 14% (5-23%); sensitivity is higher for VAB than automated CB (chi2 ((df = 1)) = 3.28, P = 0.06) but specificity is significantly higher for automated CB (14G) relative to VAB (11G) (chi2 ((df = 1)) = 6.37, P = 0.01), and the overall accuracy of the two methods is similar. Sensitivity of CNB improved with experience (over time and in relation to caseload). Accuracy was not substantially affected by lesion palpability or image-guidance method, and was similar for both masses and calcification but lower for lesions depicted as distortions on mammography. Inadequacy was very low and decreased with greater operator caseload, and was not associated with core gauge or image-guidance method. False negatives occurred in 4.4% (3.4-5.4%) of cases, and where core histology was benign but discordant with (suspicious) imaging and/or clinical findings the likelihood of malignancy was 33.1% (18.5-47.7%), emphasising the importance of correlating all test information in breast diagnosis.

Biopsy, Fine-Needle↗
PubMed2006

Sentinel node identification rate, but not accuracy, is significantly decreased after pre-operative chemotherapy in axillary node-positive breast cancer patients.

BACKGROUND: The aim was to prove the low identification rate of sentinel lymph node biopsy (SNB) and to determine the feasibility of replacing axillary lymph node dissection (AND) in axillary lymph node positive patients after chemotherapy. METHODS: From October 2001 to July 2005, 875 consecutive patients with primary operable breast cancer underwent SNB and AND. Among them, 238 received pre-operative chemotherapy. We compared the identification rate, false negative rate (FNR), negative predictive value (NPV), and accuracy of SNB in clinically node-positive patients with or without chemotherapy. RESULTS: The identification rate was significantly lower in patients received chemotherapy (77.6%) than in those not received it (97.0%) (P<0.001). In those received the therapy, the FNR was 5.6%, the NPV was 86.8%, and the accuracy was 95.9%. In those not received therapy, the FNR was 7.4% and the accuracy was 92.6% (differences not statistically significant). CONCLUSION: The identification rate in confirmed axillary lymph node-positive patients was significantly lower in patients received pre-operative chemotherapy, but accuracy did not differ significantly between the two groups. Thus, for patients who achieve complete axillary clearance by chemotherapy, SNB could replace AND.

Antineoplastic Agents↗
PubMed2005

Diagnostic accuracy of multislice computed tomography coronary angiography is improved at low heart rates.

PURPOSE: Assess the effect of heart rate on diagnostic accuracy for the detection of significant coronary artery stenosis using 16-row multislice computed tomography (MSCT). MATERIAL AND METHODS: About 120 patients (105 males; 59+/-11 years) with suspected coronary artery disease who underwent conventional coronary angiography (CA) and MSCT-CA were retrospectively enrolled for the study. Patients underwent a MSCT-CA (Sensation 16, Siemens, Germany), with the following protocol: collimation 16 x 0.75 mm, gantry rotation time 420 ms, feed/rotation 3.0 mm, kV 120, mAs 400-500. The protocol for contrast material administration was 100 ml of Iodixanol (Visipaque 320 mg l/ml, Amersham, UK) at 4 ml/s and the delay was defined with a bolus tracking technique. In all patients the mean heart rate (HR) during the scan was used as a criteria to divide the population in two groups of 60 patients each. In one group (Low HR) the 60 patients with lower heart rates, and in the other group (High HR) the patients with higher heart rates. In the two groups diagnostic accuracy (per coronary segment) for the detection of significant stenosis (>or=50% lumen reduction) was evaluated in vessels >or=2 mm of diameter using quantitative CA as reference standard. The difference in diagnostic accuracy were compared with a Chi(2) test and a p<0.05 was considered significant. RESULTS: There was no significant difference between the two groups regarding age, gender, weight, mean intravascular attenuation, and calcium score. Overall 1,310 (652 for Low HR and 658 for High HR) segments with 219 (105 for Low HR and 114 for High HR) significant lesions were available for the analysis. The average heart rate was 52+/-4 HU and 63+/-5 HU for Low HR and High HR, respectively (p<0.001). The sensitivity and specificity were 92 and 96% for Low HR and 90 and 92% for High HR (p<0.05). There were 22 vs. 44 false positives, and 8 vs. 12 false negatives in the Low HR and High HR, respectively. CONCLUSION: Increasing HR significantly deteriorates diagnostic accuracy in MSCT-CA.

Chi-Square Distribution↗
PubMed2005

Estimating the accuracy of protein structures using residual dipolar couplings.

It has been commonly recognized that residual dipolar coupling data provide a measure of quality for protein structures. To quantify this observation, a database of 100 single-domain proteins has been compiled where each protein was represented by two independently solved structures. Backbone 1H-15N dipolar couplings were simulated for the target structures and then fitted to the model structures. The fits were characterized by an R-factor which was corrected for the effects of non-uniform distribution of dipolar vectors on a unit sphere. The analyses show that favorable R values virtually guarantee high accuracy of the model structure (where accuracy is defined as the backbone coordinate rms deviation). On the other hand, unfavorable R values do not necessarily suggest low accuracy. Based on the simulated data, a simple empirical formula is proposed to estimate the accuracy of protein structures. The method is illustrated with a number of examples, including PDZ2 domain of human phosphatase hPTP1E.

Computer Simulation↗
PubMed2006

Reflections on accuracy.

The difference between test accuracy and predictive accuracy is presented and defined. The failure to distinguish between these two types of measures is shown to have led to a misguided debate over the interpretation of prevalence estimates. The distinction between test accuracy defined as sensitivity and specificity, and predictive accuracy defined as positive and negative predictive value is shown to reflect the choice of the denominator used to calculate true positive, false positive, false negative, and true negative rates. It is further shown that any instrument will tend to overestimate prevalence in low base rate populations and underestimate it in those populations where prevalence is high. The implications of these observations are then discussed in terms of the need to define diagnostic thresholds that have clinical and policy relevance.

Behavior, Addictive↗
PubMed2005

Gender and risk assessment accuracy: underestimating women's violence potential.

Understanding factors that contribute to mental health professionals' (MHPs') accuracy in assessing patients risk of violence can inform efforts to improve accuracy and to integrate risk assessment technology with practice. Based on a sample of 147 clinicians who assessed 680 patients in a psychiatric emergency room, this study investigates the influence of patient gender, MHP gender, and their potential interaction on MHPS' risk assessment accuracy. The results indicate that MHPs of both genders are particularly limited in their ability to assess female patients' risk of future violence. This finding was not limited to a particular professional group and was not attributable to gender-related differences in violence. Implications for future research on the judgment processes that may underlie MHPs' limited accuracy with women and for training programs in violence risk assessment are discussed.

Adult↗
PubMed2006

Eyewitness identification accuracy and response latency.

Eyewitness identification research has reliably shown that accurate identifications are faster than inaccurate identifications. Recently, D. Dunning and S. Perretta (2002) claimed that an identification latency of 10-12 s not only best discriminates between accurate and inaccurate identifications but also produces extremely high accuracy rates, approaching 90%. Consistent with predictions from recognition memory theory, however, we show that the optimum time boundary varies with overall response latency under manipulations of retention interval and nominal lineup size, and that the accuracy rate inside the optimum time boundary is much less impressive than previously reported. We outline directions for clarifying the accuracy and latency relationship to assist the reliable diagnosis of identification accuracy.

Analysis of Variance↗
PubMed2006

Assessment of image guided accuracy in a skull model: comparison of frameless stereotaxy techniques vs. frame-based localization.

INTRODUCTION: The use of image-guided systems (IGS) for brain biopsy has increased in neurosurgical practice. We sought to evaluate the accuracy of a plastic, disposable burr hole mounted guide for stereotactic biopsy using an IGS and compare the results of different targeting methods with those of frame based localization. METHODS: MRIs were performed on a skull model with mounted fiducials with a stereotactic frame in place and data was loaded onto the Stealth IGS. The model was placed in a Mayfield head holder and fixed to the OR table. Registration of imaging to physical space was carried out. Using three different targeting methods on the Stealth IGS, the distance between the target and the predicted position of the target, the offset error, was measured in three dimensions and confirmed by 2 observers. A sum of squares for the 3 offset errors in all planes was used to calculate the summed vector error. The same MRI dataset used with the Cosman-Roberts-Wells (CRW) stereotactic frame for comparison. The summed vector error was calculated in the same manner to compare the accuracy of targeting with these guides to the frame-based CRW system. RESULTS: For frameless stereotaxy using the "Straight- guide 4 2D" targeting method the mean error was 2.58 +/- 0.51 mm (n=12). The vector error was 5.23 +/- 0.54 (n=4). For the registration set and target using the "Offset- guide 4 2D" targeting method the mean error was 1.66 +/- 0.36 mm (n=12). The vector error was 3.32 +/- 0.72 (n=4). The best localization was obtained with the "probe's eye" planning and targeting. The mean error was 0.33 +/- 0.16 mm (n=12). The vector error was 1.0 +/- 0.28 (n=4). We found a statistical difference between the different techniques (P<0.001) (Kruskal-Wallis One Way Analysis of Variance on Ranks). An all pairwise multiple comparison procedure (Holm-Sidak method) found an overall significance level = 0.05. For the frame-based CRW the mean error from the target was 1.03 +/- 0.19 mm (n=18) and the mean target localization error vector was 2.23 +/- 0.14 (n=6). We found a statistically significant difference between NDT guide "Probes Eye" vs. the MR-CRW (P=0.003, Mann-Whitney Rank Sum Test). CONCLUSIONS: These results indicate that using MR imaging, surgical planning software and the skull mounted Navigus-DT with the probe's eye view option for targeting, localization accuracy appears to fall within acceptable ranges compared with frame-based methods which have been the standards for stereotactic brain biopsy and functional neurosurgery. Furthermore, there may be considerable differences in accuracy between different targeting methods.

Biopsy↗
PubMed2006

Diagnostic accuracy of chemical-shift MR imaging to differentiate between adrenal adenomas and non adenoma adrenal lesions.

PURPOSE: The objective of this study was to evaluate the diagnostic accuracy of chemical-shift (CS) magnetic resonance (MR) imaging in the differential diagnosis of adenoma and nonadenoma adrenal masses. MATERIALS AND METHODS: We enrolled 36 patients (9 men, 27 women, mean age 51.3+14.4 years) with unilateral (n=31) or bilateral (n=5) adrenal masses incidentally discovered during imaging examinations [ultrasound (US), computed tomography (CT)] performed for other indications. A total of 41 adrenal lesions were evaluated (mean diameter 3.0+2.2 cm). Histology (n=19), biopsy (n=3) or clinical-imaging follow-up (n=19) demonstrated 29 adenomas, five pheochromocytomas, three cysts and four carcinomas. MR imaging was performed using the following breath-hold sequences: T1-fast field echo (FFE) [repetition time (TR)/echo time (TE)=236/4.6 ms], T2-turbo spin echo-single shot (TSE-SSh) (TR/TE=831/80 ms), T1-DUAL-FFE (TR=236, double TE=4.6/2.3 ms in phase and out of phase) and T1-FFE after gadolinium-DTPA (Gd). Axial and coronal imaging planes were used, with a slice thickness of 3-5 mm. MR images were qualitatively assessed for signal intensity of the adrenal mass relative to the liver on T1, T2, CS and T1-Gd scans; diagnostic criteria for adenomas were isointensity or hypointensity on both T1 and T2 scans, out-of-phase CS signal loss and mild transient enhancement after Gd. RESULTS: Analysis of T1-T2 signal intensity showed diagnostic accuracy, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of 80%, 72%, 100%, 100% and 60%, respectively. In contrast, analysis of CS and T1-Gd signal intensity showed diagnostic accuracy, sensitivity, specificity, PPV and NPV for both sequences of 93%, 90% (p<0.05 vs. T1-T2 analysis), 100%, 100% and 80% (p<0.05 vs. T1-T2 analysis), respectively. CONCLUSIONS: CS MR imaging significantly improves characterization of adrenal masses compared with conventional T1-T2-weighted images, providing accuracy similar to that of the T1 sequence after Gd. Therefore, the CS sequence is strongly recommended for MR study of adrenal masses, and its use might obviate the need for Gd administration.

Adrenal Cortex Neoplasms↗
PubMed2026

Effect of founder breeds on genotype imputation accuracy in Canchim cattle.

UNLABELLED: Genotype imputation is a technique used to infer unobserved genotypes based on reference panels, allowing increased marker density and cost-effective optimization for genomic selection. This study aimed to evaluate whether the inclusion of genotypes from the founder breeds Nelore (NE) and Charolais (CH) improves the imputation accuracy in the composite beef cattle breed Canchim (CA). The populations studied consisted of 804 NE, 897 CH, and 392 CA animals, all genotyped using high-density panels (777,962 SNP &#x2013; single nucleotide polymorphisms). CA animals had their genotypes masked to simulate a medium-density panel (54,609 SNP). Fourteen imputation scenarios were evaluated, varying according to breed, sex, year of birth, and lineage. Imputation accuracy was determined based on the percentage of correctly imputed genotypes (PERC) and the squared Pearson&#x2019;s correlation between observed and imputed genotypes (R2). PERC values ranged from 66.52% to 97.39% and R&#xb2; from 0.6352 to 0.9780. The scenarios that included NE, CH, and CA (males or animals born before 2004) as the reference population for imputing CA females or CA animals born after 2004 showed the highest imputation accuracies. Therefore, the use of founder breeds in the reference population improves the accuracy of genotype imputation in CA cattle. The results indicate that a multibreed reference population, incorporating founder breeds, could provide a more robust and informative genetic basis for imputing composite cattle. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at https://doi.org/10.1007/s13353-026-01060-z.

Animal breeding↗
PubMed1982

Diagnostic accuracy of cardiologists compared with probability calculations using Bayes' rule.

Probability analysis has provided insights into the use of diagnostic tests in coronary artery disease, and recent developments may permit clinical application of individual patients. To validate independently two available methods of probability calculation, their diagnostic accuracy was compared with that of cardiologists. Ninety-one cardiologists participated in the study; each evaluated the clinical summaries of eight randomly selected patients. For each patient, the cardiologist assessed the probability of coronary artery disease after reviewing the clinical history, physical examination and laboratory data, including complete results from a treadmill exercise test. The probability of coronary artery disease was also obtained for each patient, using the identical information, from two methods employing Bayes' rule: (1) from a published table of data based on the patient's age, sex, symptoms and degree of S-T segment change during exercise; and (2) from a computer program using the age, sex, risk factors, resting electrocardiogram and multiple exercise measurements. Diagnostic accuracy was assessed on a scale from 0 to 100 with the coronary angiogram as the diagnostic standard. The average diagnostic accuracy on this scale was: 80.2 for the cardiologists' estimates, 78.0 for the estimates based on tables (difference from cardiologists' estimates p less than 0.05) and 83.1 for the estimates based on computer calculations (p less than 0.01). Thus probability analysis incorporating sufficient detail can achieve a diagnostic accuracy comparable with that of cardiologists. Studies of the efficacy of probability analysis in patient care are warranted.

Bayes Theorem↗
PubMed1983

Diagnostic accuracy of indium-111 platelet scintigraphy in identifying left ventricular thrombi.

This study defines the optimum imaging time window after injection of labeled platelet suspension for detection on left ventricular (LV) thrombi, identifies the most useful imaging views, and determines the reproducibility of this technique. A total of 662 images obtained from 64 patients were analyzed retrospectively on 2 separate occasions by 3 observers blinded as to patient identity, view (right anterior oblique, anterior, left anterior oblique, and left lateral), and time after injection of the platelet suspension that the images were obtained (0 to 2, 3 to 4, and 5 to 6 days). Images were categorized as either positive or negative. In every case surgical or autopsy verification of the presence or absence of LV thrombus was possible. The best combination of sensitivity, specificity, and diagnostic accuracy was found in the 3- to 4-day period in the left anterior oblique view and was 54 +/- 5% (mean +/- standard deviation), 98 +/- 1%, and 85 +/- 2%, respectively. Sensitivity, specificity, and diagnostic accuracy were not enhanced by adding additional views (right anterior oblique, left lateral, and anterior) to the left anterior oblique view in the 3- to 4-day time period. However, using multiple views, localization of thrombi to the left ventricle was facilitated. In a second retrospective analysis, a comparison of day 0 with day 3 to 4 images enhanced sensitivity and accuracy to 65 (p less than 0.001) and 90% (not significant), respectively. Specificity was unchanged at 99%. Mean intra- and interobserver agreement was 91 and 88%, respectively. Thus, (1) indium-111 platelet scintigraphy is a reproducible and specific technique for identifying LV thrombus, and (2) we advise imaging on day 0 and again 3 to 4 days after injection of the platelet suspension in right anterior oblique, left anterior oblique, left lateral, and anterior views to maximize accuracy and to facilitate localization of LV thrombus.

Blood Platelets↗