Explainable Deep Learning Improves Physician Interpretation of Myocardial Perfusion Imaging.

The accuracy of multidetector computed tomographic (CT) angiography involving 64 detectors has not been well established. We conducted a multicenter study to examine the accuracy of 64-row, 0.5-mm multidetector CT angiography as compared with conventional coronary angiography in patients with suspected coronary artery disease. Nine centers enrolled patients who underwent calcium scoring and multidetector CT angiography before conventional coronary angiography. In 291 patients with calcium scores of 600 or less, segments 1.5 mm or more in diameter were analyzed by means of CT and conventional angiography at independent core laboratories. Stenoses of 50% or more were considered obstructive. The area under the receiver-operating-characteristic curve (AUC) was used to evaluate diagnostic accuracy relative to that of conventional angiography and subsequent revascularization status, whereas disease severity was assessed with the use of the modified Duke Coronary Artery Disease Index. A total of 56% of patients had obstructive coronary artery disease. The patient-based diagnostic accuracy of quantitative CT angiography for detecting or ruling out stenoses of 50% or more according to conventional angiography revealed an AUC of 0.93 (95% confidence interval [CI], 0.90 to 0.96), with a sensitivity of 85% (95% CI, 79 to 90), a specificity of 90% (95% CI, 83 to 94), a positive predictive value of 91% (95% CI, 86 to 95), and a negative predictive value of 83% (95% CI, 75 to 89). CT angiography was similar to conventional angiography in its ability to identify patients who subsequently underwent revascularization: the AUC was 0.84 (95% CI, 0.79 to 0.88) for multidetector CT angiography and 0.82 (95% CI, 0.77 to 0.86) for conventional angiography. A per-vessel analysis of 866 vessels yielded an AUC of 0.91 (95% CI, 0.88 to 0.93). Disease severity ascertained by CT and conventional angiography was well correlated (r=0.81; 95% CI, 0.76 to 0.84). Two patients had important reactions to contrast medium after CT angiography. Multidetector CT angiography accurately identifies the presence and severity of obstructive coronary artery disease and subsequent revascularization in symptomatic patients. The negative and positive predictive values indicate that multidetector CT angiography cannot replace conventional coronary angiography at present. (ClinicalTrials.gov number, NCT00738218.)

ObjectivesCoronary computed tomography angiography (CCTA) has higher diagnostic accuracy than coronary artery calcium (CAC) score for detecting obstructive coronary artery disease (CAD) in patients with stable chest pain, while the added diagnostic value of combining CCTA with CAC is unknown. We investigated whether combining coronary CCTA with CAC score can improve the diagnosis of obstructive CAD compared with CCTA alone.MethodsA total of 2315 patients (858 women, 37%) aged 61.1 ± 10.2 from 29 original studies were included to build two CAD prediction models based on either CCTA alone or CCTA combined with the CAC score. CAD was defined as at least 50% coronary diameter stenosis on invasive coronary angiography. Models were built by using generalized linear mixed-effects models with a random intercept set for the original study. The two CAD prediction models were compared by the likelihood ratio test, while their diagnostic performance was compared using the area under the receiver-operating-characteristic curve (AUC). Net benefit (benefit of true positive versus harm of false positive) was assessed by decision curve analysis.ResultsCAD prevalence was 43.5% (1007/2315). Combining CCTA with CAC improved CAD diagnosis compared with CCTA alone (AUC: 87% [95% CI: 86 to 89%] vs. 80% [95% CI: 78 to 82%]; p < 0.001), likelihood ratio test 236.3, df: 1, p < 0.001, showing a higher net benefit across almost all threshold probabilities.ConclusionAdding the CAC score to CCTA findings in patients with stable chest pain improves the diagnostic performance in detecting CAD and the net benefit compared with CCTA alone.Clinical relevance statementCAC scoring CT performed before coronary CTA and included in the diagnostic model can improve obstructive CAD diagnosis, especially when CCTA is non-diagnostic.Key Points• The combination of coronary artery calcium with coronary computed tomography angiography showed significantly higher AUC (87%, 95% confidence interval [CI]: 86 to 89%) for diagnosis of coronary artery disease compared to coronary computed tomography angiography alone (80%, 95% CI: 78 to 82%, p < 0.001).• Diagnostic improvement was mostly seen in patients with non-diagnostic C.• The improvement in diagnostic performance and the net benefit was consistent across age groups, chest pain types, and genders.

BackgroundTreatment of sepsis requires prompt initiation of appropriate empiric antibiotics. Coverage for Gram-negative bacilli (GNB) is decided based on risk factors for resistant organisms, as documented in both structured and unstructured data in the electronic health record. We hypothesized that adding large language models (LLMs) that can decode clinical narratives would improve performance of deep learning models in predicting antimicrobial resistance in GNB sepsis.MethodsThis retrospective cohort analysis included all adult patients who met CDC sepsis criteria admitted to 10 BJC hospitals from 1/2018 to 12/2023. Sepsis episodes were classified as community-onset (< 48hrs from hospitalization) or hospital-onset. GNB were stratified as ceftriaxone-susceptible (SS), ceftriaxone-resistant but cefepime-susceptible (RS), and ceftriaxone- and cefepime-resistant (RR). Culture negative or non-GNB sepsis were labeled SS. Deep learning models (XGBoost, DeepAMR) were developed using all available structured data (e.g., demographics, vitals, labs). LLMs (BioGPT, Medical-LLaMa3-8B) were developed on full or partial history and physical exam notes. Models were assessed using area under the receiver operating characteristic curve (AUROC) and area under precision-recall curve (AUPRC).ResultsA total of 29,517 patients were included, for 23,949 community-onset and 25,101 hospital-onset sepsis episodes with RS+RR GNB in 9% and 17%, respectively (Table 1). Relevant features of structured data models included history of prior infections and resistant organisms. DeepAMR was the best performing deep learning model with AUROC 0.85/AUPRC 0.40 and AUROC 0.81/AUPRC 0.50 for community- and hospital-onset sepsis, respectively (Table 2). The best performing LLM was Medical-LLaMa3-8B on full notes (Figure 1), with AUROC 0.74/AUPRC 0.21 and AUROC 0.67/AUPRC 0.30 for community- and hospital-onset sepsis, respectively. Combining deep learning and LLMs did not significantly improve the prediction of resistant infections compared to deep learning alone.ConclusionStructured medical data provided the highest prediction of resistant GNB infections in sepsis. More analysis is needed to understand why the unstructured data did not improve model predictive performance.DisclosuresMaria Cristina Vazquez Guillamet, MD, Bionano: Stocks/Bonds (Public Company)|Charisma: Stocks/Bonds (Public Company)|Ocugen: Stocks/Bonds (Public Company)

The pathogenic mutation p.Arg14del in the gene encoding Phospholamban (PLN) is known to cause cardiomyopathy and leads to increased risk of sudden cardiac death. Automatic tools might improve the detection of patients with this rare disease. Deep learning is currently the state-of-the-art in signal processing but requires large amounts of data to train the algorithms. In situations with relatively small amounts of data, like PLN, transfer learning may improve accuracy. We propose an ECG-based detection of the PLN mutation using transfer learning from a model originally trained for sex identification. The sex identification model was trained with 256,278 ECGs and subsequently finetuned for PLN detection (155 ECGs of patients with PLN) with two control groups: a balanced age/sex matched group and a randomly selected imbalanced population. The data was split in 10 folds and 20% of the training data was used for validation and early stopping. The models were evaluated with the area under the receiver operating characteristic curve (AUROC) of the testing data. We used gradient activation for explanation of the prediction models. The models trained with transfer learning outperformed the models trained from scratch for both the balanced (AUROC 0.87 vs AUROC 0.71) and imbalanced (AUROC 0.0.90 vs AUROC 0.65) population. The proposed approach was able to improve the accuracy of a rare disease detection model by transfer learning information from a non-manual annotated and abundant label with only limited data available.

Breast cancer is the most common cancer worldwide, and magnetic resonance imaging (MRI) constitutes a very sensitive technique for invasive cancer detection. When reviewing breast MRI examination, clinical radiologists rely on multimodal information, composed of imaging data but also information not present in the images such as clinical information. Most machine learning (ML) approaches are not well suited for multimodal data. However, attention-based architectures, such as Transformers, are flexible and therefore good candidates for integrating multimodal data. The aim of this study was to develop and evaluate a novel multimodal deep learning (DL) model combining ultrafast dynamic contrast-enhanced (UF-DCE) MRI images, lesion characteristics and clinical information for breast lesion classification. From 2019 to 2023, UF-DCE breast images and radiology reports of 240 patients were retrospectively collected from a single clinical center and annotated. Imaging data were constituted of volumes of interest (VOI) extracted around segmented lesions. Non-imaging data were constituted of both clinical (categorical) and geometrical (scalar) data. Clinical data were extracted from annotated reports and were associated to their corresponding lesions. We compared the diagnostic performances of traditional ML methods for non-imaging data, an image model based on the DL architecture, and a novel Transformer-based architecture, the Multimodal Sieve Transformer with Vision Transformer encoder (MMST-V). The final dataset included 987 lesions (280 benign, 121 malignant lesions, and 586 benign lymph nodes) and 1081 reports. For classification with non-imaging data, scalar data had a greater influence on performances of lesion classification (Area under the receiver operating characteristic curve (AUROC)=0.875±0.042) than categorical data (AUROC=0.680±0.060). MMST-V achieved better performances (AUROC=0.928±0.027) than classification based on non-imaging data (AUROC=0.900±0.045), and imaging data only (AUROC=0.863±0.025). The proposed MMST-V is an adaptative approach that can consider redundant information provided by multimodal information. It demonstrated better performances than unimodal methods. Results highlight that the combination of clinical patient data and detailed lesion information as additional clinical knowledge enhances the diagnostic performances of UF-DCE breast MRI.

Aim to evaluate the relationships between functional and anatomical information obtained by myocardial perfusion imaging (MPI) and coronary computed tomography angiography (CCTA) in a series of consecutive patients at intermediate probability of coronary artery disease (CAD). Material and Methods — The study group comprised 139 patients (83 men, age of 61.6±7.5 years) who underwent CCTA and single-photon emission computed tomography myocardial perfusion imaging (SPECT-MPI). Based on CCTA results patients were divided into three groups: 1) with the absence of coronary atherosclerosis on CCTA; 2) with non-obstructive CAD (&lt;50%); 3) with obstructive (≥50%) CAD. The Segment Involvement Score, Segment Stenosis Score (SSS) and CTA Risk Score were calculated as measures of global atherosclerosis burden. MPI studies were considered abnormal in the presence of SSS≥4. Results — Abnormal myocardial perfusion was detected in 60% of cases in group 1 and 2; in 75% of cases in group 3. The overall frequencies of normal and abnormal MPI studies differed significantly only in obstructive CAD patients and did not differ in group 1 and 2. There were no significant correlations between calcium score, atherosclerotic lesion length, positive remodelling index and MPI results in patients with non-obstructive as well as in patients with obstructive CAD. In group of patients with obstructive CAD Segment Stenosis Score correlated wekly with SSS (r=0.39, p=0.001) and SDS (r=0.28; p=0.012); the CTA Risk Score showed correlationes with SSS (r=0.38, p=0.002) and SDS (r=0.30, p=0.020). Conclusion — Myocardial perfusion abnormalities may develop even in the absence of critical coronary artery lesions. The extent of myocardial ischemia correlates with measures of global CAD burden only in patients with obstructive CAD.

Metabolic dysfunction-associated steatotic liver disease (MASLD) can progress to metabolic dysfunction-associated steatohepatitis (MASH) and liver fibrosis, contributing to a heavier global health burden. Non-invasive diagnostic tools developed using machine learning (ML) and deep learning (DL), two representative artificial intelligence algorithms, are increasingly being explored for MASH and its related fibrosis assessment. This study aimed to compare the diagnostic performance of different ML and DL models and identify the top-performing models for diagnosing MASH and associated liver fibrosis. A systematic review and meta-analysis were conducted across PubMed, Web of Science, Embase and Cochrane Library from inception to May 18, 2025. Pooled area under the receiver operator characteristic curve (AUROC) values with 95 % confidence interval (CI) were calculated. Accuracy, specificity, sensitivity, positive predictive values, and negative predictive values were also recorded. Of 4,314 studies initially identified, 106 met the inclusion criteria, with 35 studies (ML: n=28; DL: n=7) providing data for analysis. Logistic Regression and Neural Network are the most commonly algorithms applied in ML and DL, respectively. The pooled AUROCs for diagnosing MASH were 0.833 (95 %CI: 0.806-0.860) for ML models and 0.841 (95 %CI: 0.782-0.900) for DL models. Light Gradient Boosting Machine (LightGBM) and ResNet50 were the best-performing models for diagnosing MASH within ML and DL algorithms, respectively, achieving corresponding AUROCs of 0.920 (95 %CI: 0.916-0.924) and 0.960 (95 %CI: 0.951-0.969). For fibrosis diagnosis, ML models had a pooled AUROC of 0.826 (95 %CI: 0.792-0.860), with Categorical Boosting (CatBoost) achieving the highest AUROC of 0.960 (95 %CI: 0.950-0.970). DL models yielded the pooled AUROC of 0.875 (95 %CI: 0.816-0.934) for fibrosis diagnosis. Both ML and DL models demonstrated strong diagnostic performance for MASH and liver fibrosis, with DL achieving marginally higher AUROCs. AI-driven approaches show promise in MASLD management.

ObjectivePrevious studies have shown that global coronary artery calcium score (CACS) can improve single photon emission computerized tomography (SPECT) myocardial perfusion imaging (MPI) to detect obstructive coronary artery disease (CAD). Whether regional CACS can improve SPECT MPI to detect obstructive CAD remains unclear. The aim of this study was to verify whether regional CACS has additional diagnostic value for obstructive CAD in suspected patients, compared to SPECT MPI and global CACS.MethodsThe study included 321 suspected CAD patients who underwent one-stop rest-stress SPECT MPI and low-dose computed tomography (CT) scan. All patients underwent coronary angiography within one month after examination. MPI images were visually analyzed by 2 experienced nuclear cardiologists. The regional CACS of left anterior descending coronary artery (LAD), left circumflex coronary artery (LCX), right coronary artery (RCA) and global CACS were calculated. Obstructive CAD was defined as ≥ 70% narrowing of the inner diameter of the LAD, LCX, RCA or their main branches and ≥ 50% narrowing of the left main coronary artery (LM).ResultsAmong the 321 patients, 86 (26.8%, 86/321) had obstructive CAD. With the increased in global and regional CACS, there was an increasing trend of patients with obstructive CAD (P for trend < 0.001). Regional CACS had a better diagnostic performance in RCA territories (AUC 0.856, P < 0.001) compared with LAD, LCX territories (AUC 0.690, 0.674, respectively). The AUC of combined regional CACS and MPI was significantly higher than that of MPI alone (0.735 vs. 0.600, P < 0.001). However, based on MPI, the AUC of combined regional CACS was not significantly higher than that of global CACS (0.735 vs. 0.732, P = 0.898). The sensitivity and specificity of regional CACS combined with MPI for detecting obstructive CAD were 64.0% and 72.8%, respectively.ConclusionsRegional CACS was effective in detecting obstructive CAD in RCA territory. Based on SPECT MPI, regional CACS improved the detection of obstructive CAD, but was not superior to global CACS.

In this multicentric study, individual patient data from 3 different centers were analyzed. Consecutive patients receiving standardized multiparametric breast magnetic resonance imaging for standard nonscreening indications were included. At each center, 2 experienced radiologists with more than 5 years of experience retrospectively interpreted the examinations in consensus and applied the KS to every histologically verified lesion. The corresponding mean ADC of each lesion was measured using a Wielema type 4 region of interest. According to established methods, the KS and ADC were combined, yielding the KS+ score. Diagnostic accuracy was evaluated by the area under the receiver operating characteristics curve (AUROC) and compared between the KS, ADC, and KS+ (DeLong test). Likewise, the potential to help avoid unnecessary biopsies was compared between the KS, ADC, and KS+ based on established high sensitivity thresholds (McNemar test). A total of 450 lesions in 414 patients (mean age, 51.5 years; interquartile range, 42-60.8 years) were included, with 219 lesions being malignant (48.7%; 95% confidence interval [CI], 44%-53.4%). The performance of the KS (AUROC, 0.915; CI, 0.886-0.939) was significantly better than that of the ADC (AUROC, 0.848; CI, 0.811-0.880; P < 0.001). The largest difference between these parameters was observed when assessing subcentimeter lesions (AUROC, 0.909 for KS; CI, 0.849-0.950 vs 0.811 for ADC; CI, 0.737-0.871; P = 0.02).The use of the KS+ (AUROC, 0.918; CI, 0.889-0.942) improved the performance slightly, but without any significant difference relative to a single KS or ADC reading (P = 0.64).When applying high sensitivity thresholds for avoiding unnecessary biopsies, the KS and ADC achieved equal sensitivity (97.7% for both; cutoff values, >4 for KS and ≤1.4 × 10-3 mm2/s for ADC). However, the rate of potentially avoidable biopsies was higher when using the KS (specificity: 65.4% for KS vs 32.9% for ADC; P < 0.0001). The KS was superior to the KS+ in avoiding unnecessary biopsies. Both the KS and ADC may be used to distinguish benign from malignant breast lesions. However, KS proved superior in this task including, most of all, when assessing small lesions less than 1 cm. Using the KS may avoid twice as many unnecessary biopsies, and the combination of both the KS and ADS does not improve diagnostic performance.

The influence of coronary calcium score on the interpretation of myocardial perfusion imaging

Identification of patients with acute coronary syndrome requiring coronary revascularization can be challenging due to inconclusive electrocardiogram (ECG) findings or biomarker results. A deep learning model to detect ECG patterns associated with revascularization likelihood was developed, aiming to guide further assessment and reduce diagnostic uncertainty. A convolutional neural network model was trained on 144 691 ED visits from a US cohort (60 ± 19 years; 53% female; 0.6% revascularization), tested on a separate test cohort (n = 35 995), and benchmarked against clinician ECG interpretation and cardiac troponin T (TnT). External validation was performed for the outcomes revascularization and Type 1 myocardial infarction (MI) on 18 673 ED visits from Europe (55 ± 21 years; 49% female; 1.5% revascularization; 1% Type 1 MI). Primary performance metric was area under the receiver operating characteristic curve (AUROC). In the test cohort, the model achieved an AUROC of 0.91 (95% confidence interval [CI] 0.91-0.91), outperforming clinician ECG interpretation (AUROC 0.65, 95% CI 0.54-0.76) and conventional cardiac TnT (AUROC 0.71). In the external validation cohort, ECG model AUROC was 0.81 (95% CI 0.81-0.82) for revascularization, and 0.85 (95% CI 0.84-0.85) for Type 1 MI, compared with 0.70 (95% CI 0.57-0.83) and 0.74 (95% CI 0.56-0.92) for clinician interpretation, and 0.85 and 0.87 for high-sensitivity (hs)-TnT, respectively. The ECG model had higher specificity but lower sensitivity compared with high-sensitivity-troponin T. The model was able to detect revascularization and Type 1 MI with competitive performance, suggesting a potential role to complement current clinical assessment.

Coronary artery disease (CAD) results in substantial morbidity and mortality. The purpose of this study was to develop a deep learning model to detect CAD defined using diagnostic codes ("ECG2CAD") and identify people at risk for adverse events using electrocardiograms (ECGs) in a primary care setting. ECG2CAD was trained on 764,670 ECGs representing 137,199 individuals at Massachusetts General Hospital (MGH). Model performance for discrimination of prevalent CAD was measured using area under the receiver operating characteristic curve (AUROC) and area under the precision-recall curve (AUPRC), and compared against model of age and sex, and Pooled Cohort Equations, in 3 test sets: MGH, Brigham and Women's Hospital (BWH), and UK Biobank. Subgroups were assessed for incident CAD-related events in a BWH primary care cohort. ECG2CAD was evaluated in MGH (N = 18,706 [6,051 cases], age 57 ± 16 years), BWH (N = 88,270 [27,898 cases], age 57 ± 16 years), and UK Biobank (N = 42,147 [1,509 cases], age 65 ± 8 years). ECG2CAD consistently discriminated prevalent CAD (MGH AUROC: 0.782; AUPRC: 0.639; BWH: AUROC: 0.747; AUPRC: 0.588; UK Biobank AUROC: 0.760; AUPRC: 0.155) and incrementally vs models based on age and sex or Pooled Cohort Equations (P < 0.01) in MGH and BWH. In the BWH primary care subset, model performance was consistent across subgroups. Being in the highest quintile of ECG2CAD risk was associated with higher risk for adverse events compared with low-risk group (myocardial infarction HR: 5.59; 95% CI: 4.76-6.56, heart failure 10.49; 95% CI: 7.96-13.84, all-cause mortality 2.68; 95% CI: 2.32-3.10). Artificial intelligence-enabled analysis of the ECG may facilitate identification of individuals with possible undiagnosed CAD and inform downstream testing and preventive measures.

Circulating miRNAs previously associated with androgen excess in women might be used as diagnostic biomarkers for polycystic ovary syndrome (PCOS). Models based on circulating miR-142-3p and miR-598-3p expression show good discrimination among women with and without PCOS, particularly when coupled with easily available measurements such as waist-to-hip ratio (WHR) and circulating LH-to-FSH (LH/FSH) ratios. The lack of standardization of the signs, methods, and threshold values used to establish the presence of the diagnostic criteria (hyperandrogenism, ovulatory dysfunction, and polycystic ovarian morphology) complicates the diagnosis of PCOS. Certain biomarkers may help with such a diagnosis. We conducted a validation study to check the diagnostic accuracy for PCOS of several miRNAs that were associated with the syndrome in a small pilot study that had been previously carried out by our research group. This was a diagnostic test study involving 140 premenopausal women. We included 71 women with PCOS and 69 healthy control women in the study. Both groups were selected as to be similar in terms of body mass index. We used miRCURY LNA™ Universal RT microRNA PCR to analyse the five miRNAs that had shown the strongest associations with PCOS in a much smaller pilot study previously conducted by our group. We studied diagnostic accuracy using receiver operating characteristics (ROC) curve analysis. Only the expression of two miRNAs, miR-142-3p and miR-598-3p, of the five studied, was different between the women with PCOS and the non-hyperandrogenic controls. The diagnostic accuracy of the combination of these circulating miRNAs was good (area under the ROC curve (AUC) 0.801; 95% CI: 0.72-0.88) and was further improved when adding WHR (AUC 0.834, 95% CI: 0.756-0.912), LH/FSH ratio (AUC = 0.869, 95% CI: 0.804-0.934) or both (AUC = 0.895, 95% CI: 0.835-0.954). We developed several models by selecting different threshold values for these variables favouring either sensitivity or specificity, with positive and negative predictive values as high as 88% or 85%, respectively. Patients included here had the classic PCOS phenotype, consisting of hyperandrogenism and ovulatory dysfunction; hence, the present results might not apply to milder phenotypes lacking androgen excess. If confirmed in larger studies addressing different populations and PCOS phenotypes, these biomarkers may be useful to simplify the clinical diagnosis of this prevalent syndrome. This research was funded by Instituto de Salud Carlos III, Spanish Ministry of Science and Innovation (grants PI15/01686, PIE16/00050, PI18/01122 & PI21/00116) and co-funded by European Regional Development Fund 'A way to make Europe'. Centro de Investigación Biomédica en Red Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM) and Instituto Ramón y Cajal de Investigación Sanitaria (IRYCIS) are also initiatives of the Instituto de Salud Carlos III. The authors have no competing interests to disclose. N/A.

Phosphatidylethanol and ethyl glucuronide to categorize alcohol consumption in alcohol-related cirrhosis.

The principal goal of this study was to compare the diagnostic accuracy of three brief instruments in memory clinic attendees. Two of the instruments were based on face-to-face clinical assessment (Standardized Mini-mental State Examination and Rowland Universal Dementia Assessment Scale), whereas the third group used proxy information from an informant (Informant Questionnaire on Cognitive Decline in the Elderly). Dementia diagnosis as provided by a specialist physician (geriatrician, psychiatrist or neurologist) was used as the reference standard. This was a cross-sectional study. Data were collected from 204 consecutive memory clinic attendees (M = 76.90, 56% female) and their family caregivers. Comparative utility was assessed through receiver operating characteristic (ROC) analyses. One hundred and fifty-two patients (75%) were diagnosed as having dementia. Diagnostic accuracy, as indicated by the area under the ROC curve (AUC), was similar for the three instruments as follows: SMMSE (AUC = 0.82, 95% CI = 0.76, 0.87, p < 0.0001) and RUDAS (AUC = 0.83, 95% CI = 0.77, 0.88, p < 0.0001), and slightly lower for IQCODE (AUC = 0.77, 95% CI = 0.71, 0.83, p < 0.0001). There was no significant difference between the areas under the curve (χ2 = 2.57, df = 2, p = 0.28). Diagnostic accuracy was similar for the three instruments, which all proved to be moderately useful tools for initial screening for cognitive impairment in the memory clinic environment. Being a proxy measure, the IQCODE had specific practical use in this context, where the patient might not be able to provide information. The RUDAS exhibited high specificity and proved to be less dependent upon cultural factors than the SMMSE, making it particularly valuable in a multicultural setting.