True Negatives Research Articles

Analysing the impact of size of NGS panel in defining first line therapeutic strategies in NSCLC.

3070 Background: NCCN recommends the analysis of 8 genes (EGFR, ALK, ROS1, BRAF, KRAS, MET, RET, ERBB2, and NTRK1/2/3) for NSCLC patients to identify efficacious target therapies. Concerned with the rising incidence of sub-optimal response to first-line therapy and rather early progression of disease we performed retrospective analysis in a subset of patients treated at our hospital in order to streamline molecular evaluation strategies. Methods: In this study, we retrospectively evaluated the impact of NGS panel sizes in therapy-naïve NSCLC patients. 242 therapy naïve patients evaluated for molecular genetic profiling were stratified into three groups based on gene panel size: a) Small panel (<20 genes): Focused on NCCN-recommended genes, b) Medium panel (50–100 genes): Included organ agnostic genes, c) Comprehensive panel (>100 genes): Included genomic signatures like TMB, MSI & HRD scores. Results: Of 242 therapy-naïve NSCLC patients, 60% (145/242) were evaluated using a small panel of which 13% (19/145) had no detectable genetic alterations while 37% (54/145) had 1 st line targetable mutations, 31% (45/145) exhibited both targetable and resistance causing mutations, and 19% (28/145) showed only resistance causing mutations. In the 50–100 genes Panel, comprising 29% (70/242) of patients, 10% (7/70) had no genetic alterations, while 26% (18/70) had 1 st line targetable mutations, 30% (21/70) demonstrated both targetable and resistance mutations, and 34% (24/70) harboured only resistance causing mutations. Finally, in the comprehensive NGS group (>100 genes), which accounted for 11% (25/242) of cases, only 4% (1/25) lacked detectable genetic alterations; while, 12% (3/25) had 1 st line targetable mutations, 32% (8/25) exhibited both targetable and resistance causing mutations, and 52% (13/25) showed only resistance causing mutations. Conclusions: a. Increase in gene panel size results in reduction of true negatives. Hence smaller panels may not necessarily capture resistance causing mutations. b. As the gene panel size increases, the detection of actionable driver mutations (e.g., EGFR, ALK) remains consistent; however, there is a notable shift in the mutation profile, with a decrease in cases harbouring only targetable mutations and an increase in those exhibiting both actionable and resistance causing mutations. Hence opting for comprehensive NGS profiling at baseline may increase diagnostic costs marginally, but will have significant impact in designing more effective 1 st line therapeutic strategies.

Smart neural network and cognitive computing process for multi task nuclei detection segmentation and classification in breast cancer histopathology images

The detection, segmentation, and differentiation of benign and malignant nuclei from the histopathology images is a challenging task for the early diagnosis of breast cancer. Misinterpretation of True Negative (TN) and False Positive (FP) can generate incorrect results. The proposed Cognitive Computing Process (CCP) detects and segments the nuclei using Deep U-Net with Spatial Attention Mechanisms (SAM) and microns-per-pixel measurements to accurately locate and assess nuclei density. To separate the nuclei of benign and malignant, the patches are introduced to leverage the model’s learning process. The proposed Smart Neural Network (SNN) models contain Smart Convolutional Neural Network (SCNN) and Deep Convolutional Neural Network (DCNN) to reduce incorrect results. Proposed CCP and SNN were evaluated using the BreakHis dataset, which contains 5547 images of benign and malignant samples at various magnifications (40×, 100×, 200×, 400×). These images were processed into patches, totaling 11,642, 9282, 9102, and 9678 patches, each 224 × 224 pixels. The CCP model outperformed state-of-the-art models UNet, Residual UNet (ResUNet), and Convolutional Neural Network Long Short-Term Memory (CNN-LSTM) with a Dice coefficient of 99.90%, an F1-score of 99.04%, a precision of 99.80%, and a recall of 99.76%. The learning process began with a learning rate of 0.01 and a decay rate of 0.8, and the SCNN achieved false negative and false positive rates of 0.04 and 0.05 for low-density nuclei at 400× and 40× magnification, respectively. In contrast, the Deep Convolutional Neural Network (DCNN) recorded rates of 0.02 and 0.01. For high-density patches, the SCNN model FN and FP rates of 0.0 and 0.08, while the DCNN reported 0.09 and 0.0. The proposed learning process with Smart Neural Networks (SNN) achieved high precision (77–99%), recall (75–99%), F1-score (75–99%), and an AUC of 86–100%. The combination of CCP and SNN improved accuracy over existing CNN models like ResNet50, VGG19, DenseNet109, DenseNet201, and VGG16. An ablation study showed a p-value of 0.00003 based on the AUC, highlighting the model’s potential to enhance automated breast cancer diagnosis and support clinical decision-making.

Quality over quantity: how to get the best results when using docking for repurposing

Molecular docking is among the fastest and most readily available computational tools to explore protein–ligand interactions. However, little effort has been put into assessing the quality of its results. In this paper, we compared eight free license docking programs to screen a drug library against the human target, phosphodiesterase 5A (PDE5A), to evaluate their ability to find its known ligand, sildenafil, and other ligands that became erectile dysfunction drugs because they inhibit this target. GNINA was superior at identifying the known target because it offers a convolutional neural network (CNN) score that ranks the quality of docking results. Using this CNN score improved the ranking of known positives. Receiver operating characteristic (ROC) analysis revealed that all docking suites lack specificity; that is, they often misidentify true negatives. Employing a CNN score cutoff before ranking by docking affinity raised specificity with a small loss in sensitivity. After the cutoff, datasets became smaller but of higher quality. We propose a heuristic to produce relevant docking results, which includes an overall evaluation of the target on docking performance through ROC and an improvement of candidate binder selection using a CNN score cutoff of 0.9.

Clinical Validation and Post-Implementation Performance Monitoring of a Neural Network-Assisted Approach for Detecting Chronic Lymphocytic Leukemia Minimal Residual Disease by Flow Cytometry.

Background: Flow cytometric detection of minimal residual disease (MRD) in chronic lymphocytic leukemia (CLL) is complex, time-consuming, and subject to inter-operator variability. Deep neural networks (DNNs) offer potential for standardization and efficiency improvement, but require rigorous validation and monitoring for safe clinical implementation. Methods: We evaluated a DNN-assisted human-in-the-loop approach for CLL MRD detection. Initial validation included method comparison against manual analysis (n = 240), precision studies, and analytical sensitivity verification. Post-implementation monitoring comprised four components: daily electronic quality control, input data drift detection, error analysis, and attribute acceptance sampling. Laboratory efficiency was assessed through a timing study of 161 cases analyzed by five technologists. Results: Method comparison demonstrated 97.5% concordance with manual analysis for qualitative classification (sensitivity 100%, specificity 95%) and excellent correlation for quantitative assessment (r = 0.99, Deming slope = 0.99). Precision studies confirmed high repeatability and within-laboratory precision across multiple operators. Analytical sensitivity was verified at 0.002% MRD. Post-implementation monitoring identified 2.97% of cases (26/874) with input data drift, primarily high-burden CLL and non-CLL neoplasms. Error analysis showed the DNN alone achieved 97% sensitivity compared to human-in-the-loop-reviewed results, with 13 missed cases (1.5%) showing atypical immunophenotypes. Attribute acceptance sampling confirmed 98.8% of reported negative cases were true negatives. The DNN-assisted workflow reduced average analysis time by 60.3% compared to manual analysis (4.2 ± 2.3 vs. 10.5 ± 5.8 min). Conclusions: The implementation of a DNN-assisted approach for CLL MRD detection in a clinical laboratory provides diagnostic performance equivalent to expert manual analysis while substantially reducing analysis time. Comprehensive performance monitoring ensures ongoing safety and effectiveness in routine clinical practice. This approach provides a model for responsible AI integration in clinical laboratories, balancing automation benefits with expert oversight.

A systematic review and meta-analysis of the diagnostic accuracy after preimplantation genetic testing for aneuploidy.

Aneuploidy accounts for many pregnancy failures and congenital anomalies. Preimplantation genetic testing for aneuploidy (PGT-A) is a screening test applied to embryos created from in vitro fertilization to diminish the chance of an aneuploid conception. The rate of misdiagnosis for both false aneuploidy (false positive) and false euploidy (false negative) test results is unknown. The objective of this study was to determine the rate of misclassification of both aneuploidy and euploidy after PGT-A. We conducted a systematic review and meta-analysis. We searched Medline, Embase, Cochrane Central, CINAHL and WHO Clinical Trials Registry from inception until April 10, 2024. The protocol was registered in International Prospective Register of Systematic Reviews (PROSPERO CRD 42020219074). We included studies that conducted either a pre-clinical validation of the genetic platform for PGT-A using a cell line, studies that compared the embryo biopsy results to those from the whole dissected embryo or its inner cell mass (WE/ICM), and studies that compared the biopsy results to prenatal or postnatal genetic testing. Two independent reviewers extracted true and false positives and negatives comparing biopsy results to the reference standard (known karyotype, WE/ICM, pregnancy outcome). For preclinical studies, the main outcome was the positive and negative predictive values. Misdiagnosis rate was the outcome for pregnancy outcome studies. The electronic search yielded 6674 citations, of which 109 were included. For WE/ICM studies (n=40), PPV was 89.2% (95% CI 83.1-94.0) and NPV was 94.2% (95% CI 91.1-96.7, I2=42%) for aneuploid and euploid embryos, respectively. The PPV for mosaic embryos of either a confirmatory mosaic or aneuploid result was 52.8% (95% CI 37.9-67.5). For pregnancy outcome studies (n=43), the misdiagnosis rate after euploid embryo transfer was 0.2% (95% CI 0.0-0.7%, I2=65%). However, the rate for mosaic transfer, with a confirmatory euploid pregnancy outcome, was 21.7% (95% CI: 9.6-36.9, I2=95%). The accuracy of an aneuploid result from PGT-A is excellent and can be relied upon as a screening tool for embryos to avoid aneuploid pregnancies. Similarly, the misdiagnosis rate after euploid embryo transfer is less than 1%. However, there is a significant limitation in the accuracy of mosaic embryos.

Disposable Sensor Array Embedded in Facemasks for the Identification of Chronic Kidney Disease.

The increasing global burden of chronic kidney disease (CKD) necessitates the development of simple and inexpensive diagnostic tools. Capitalizing on the relationship between breath composition and CKD, we introduce a disposable array of four resistive gas sensors printed on a low-cost, disposable substrate and embedded in the internal layers of FFP2 facemasks. Sensors are based on blends of porphyrins─a molecular family often used in breath analysis─and the PEDOT/PSS conducting polymer. The individual sensors demonstrate remarkable sensitivity to ammonia and other CKD-related volatile compounds, while the combinatorial selectivity of the sensor array enables the identification of volatile compounds regardless of their concentration. The diagnostic capabilities of the device were tested on a cohort of CKD patients and a control group. To address the absence of a reference gas inside the facemask, we developed a measurement protocol based on breathing cycles at different rates. The application of a continuous wavelet transform to the sensor signals produces stable and reproducible features. Linear Discriminant Analysis of sensor features achieved the identification of CKD patients with 93.3% true positives and 86.7% true negatives. Additional evidence suggests that the sensor array can stratify CKD patients according to the severity of renal dysfunction, indicating its potential use in monitoring disease progression.

Discrimination of a single-item scale to measure intention to have a COVID-19 vaccine.

When developing public health measures in a pandemic, it is important to examine attitudes and beliefs relating to vaccination uptake. We report the discrimination of a single-item vaccination intention scale and derive cutpoints in terms of sensitivity (true positives) and specificity (true negatives) in relation to subsequent vaccination status. In a sample of UK adults (n=1119) recruited through an online survey platform, vaccination intention was measured on a 0-10 numerical rating scale (0=very unlikely, 10=very likely) at the beginning of the UK COVID-19 vaccination rollout (January 2021), and self-reported vaccination status was gathered after vaccination had been offered to all adults (October 2021). Discrimination of the scale was measured by the area under the receiver operating characteristic (ROC) curve. The responders reporting being vaccinated or unvaccinated were 1034 (92.4%) and 85 (7.6%), respectively. The area under the ROC curve was.956 (95% CI.943,.967), indicating a high degree of discrimination. The combined value of sensitivity and specificity was greatest at a cutpoint of 8 on the scale (sensitivity =.821, specificity =.988). If, however, the individual values of sensitivity and specificity are required to be simultaneously optimized, this occurs at point 6 (sensitivity =.886, specificity =.871). We recommend a 0-10 intention scale as a validated, practical measure of vaccination intention in public health practice, with a cutpoint of 8 on the scale as optimal, unless sensitivity and specificity are to be simultaneously optimized, when 6 is the optimal cutpoint.

Diagnostic accuracy of Ottawa rules in diagnosing ankle fractures among patients taking X ray as gold standard.

Objective: To determine the diagnostic accuracy of the Ottawa Ankle Rules (OAR) in diagnosing ankle fractures among patients, with X-ray imaging as the gold standard. Study Design: Cross-sectional study. Setting: Emergency Department of Lady Reading Hospital, Peshawar. Period: 1st January 2024 to 30th June 2024. Methods: 286 cases patients aged 18 to 60 years, presenting with ankle twisting and pain within 6 hours, were included. Exclusion criteria included patients unable to answer the Ottawa questionnaire or those refusing X-ray imaging. Following informed consent, patients were assessed using the Ottawa Ankle Rules and underwent X-ray imaging. The results were classified into true positives, false positives, true negatives, and false negatives. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy were calculated. Results: The mean age of participants was 36.73±6.7 years, with 67.13% males and 32.86% females. Ankle fractures were more common on the right side (62.93%). Among normal radiographs, 55.24% were correctly classified, while 18.18% were false positives. In patients with radiographic fractures, 23.77% were correctly identified. The sensitivity of the Ottawa Rules was 95.18%, while specificity was 56.67%. The positive predictive value was 68.72%, and the negative predictive value was 92.16%. Sensitivity was high in both males (92.73%) and females (94.55%), though specificity was lower in males (42.86%) compared to females (56.10%). Conclusion: The Ottawa Ankle Rules demonstrated high sensitivity for detecting ankle fractures but lower specificity.

Accuracy and reproducibility of latent print decisions on comparisons from searches of an automated fingerprint identification system.

Accuracy and reproducibility of latent print decisions on comparisons from searches of an automated fingerprint identification system.

Extemporaneous histological analysis according to slow-Mohs combined with Full-Field Optical Coherence Tomography evaluation (FFOCT) in cutaneous tumor pathology: Toward a digital extemporaneous analysis?

Extemporaneous histological analysis according to slow-Mohs combined with Full-Field Optical Coherence Tomography evaluation (FFOCT) in cutaneous tumor pathology: Toward a digital extemporaneous analysis?

Real-world accuracy of SARS-CoV-2 antigen detection compared with qPCR: A cross-sectional study in Toledo - PR, Brazil.

Real-world accuracy of SARS-CoV-2 antigen detection compared with qPCR: A cross-sectional study in Toledo - PR, Brazil.

Clinical assessment of the criticality index - dynamic, a machine learning prediction model of future care needs in pediatric inpatients.

To assess patient characteristics and care factors that are associated with correct and incorrect predictions of future care locations (ICU vs. non-ICU) by the Criticality Index-Dynamic (CI-D), with the goal of enhancing the CI-D. Retrospective structured chart review. All pediatric inpatients admitted from January` 1st 2018 - February 29th 2020 through the emergency department. Patient characteristics and care factors associated with correct (true positives, true negatives) and incorrect predictions (false positives, false negatives) of future care locations (ICU vs. non-ICU) by the CI-D were assessed. Of the 3,018, patients, 139 transitioned from non-ICU locations to ICU care; 482 were transferred from the ICU to non-ICU care locations, and 2,400 remained in non-ICU care locations. For the ICU Prediction group, the false negative patients were older, more frequently male, and had longer hospital and ICU lengths of stay compared to the true positive patients. The significant differences in the ICU Prediction group for false negative compared to the true positive patients included a less frequent: primary diagnosis of respiratory failure, use of high flow nasal canula, hourly cardio-respiratory vital signs prior to transfer to the ICU, and neurologic vital signs after transfer from the ICU. For the ICU Discharge prediction group, false positive patients were more frequently: younger, had a primary diagnosis of respiratory failure, more frequently received respiratory support after discharge from the ICU, and received less frequent neurological vital signs prior to transfer from the ICU. For the Non-transfer prediction category, demographics and clinical variables did not differ between the true negative and false positive prediction groups. We conducted the first comprehensive analysis via structured chart reviews of patient characteristics and care factors that are associated with correct and incorrect predictions of future care locations by the machine learning algorithm, the CI-D, gaining insights into potential new predictor variables for inclusion in the model to improve future model iterations.

Histopathology is More Reliable Than Microbiology for Detecting Residual Osteomyelitis After Conservative Surgery for Diabetic Foot: The Pitfall of False-Positive Cultures and the Role of Pseudomonas aeruginosa.

The optimal method for assessing residual osteomyelitis after conservative surgery for diabetic foot infection remains controversial. Microbiological cultures are frequently used due to their rapid turnaround and utility in guiding antibiotic therapy, but their diagnostic reliability is uncertain. This study compared microbiological cultures and histopathology in evaluating bone resection margins, using histopathology as the gold standard. We included 93 patients undergoing conservative surgery for diabetic foot osteomyelitis. Bone samples were obtained from the proximal resection margin for both microbiology and histopathology. A microbiological result was considered contamination when cultures were positive but histopathology was negative. Microbiological cultures at bone resection margins yielded 52 true positives, 23 false positives (contamination), 10 false negatives, and 8 true negatives when compared to histopathology. This resulted in a sensitivity of 83.9%, specificity of 25.8%, positive predictive value of 69.3%, and negative predictive value of 44.4%. Contamination was not associated with the severity of infection, presence of soft tissue involvement, inflammatory markers, or glycemic control. No association was found between contamination and polymicrobial flora in the primary surgical specimen. However, Pseudomonas aeruginosa was the only species significantly associated with contamination (p = .008), suggesting species-specific factors may contribute to microbiological false positives. These findings emphasize that microbiology, while sensitive, is not a specific method for assessing residual bone infection and should not be used in isolation. Histopathology remains the more reliable diagnostic tool. Future research should explore biofilm-targeted strategies and intraoperative antiseptic protocols to reduce contamination of bone biopsy specimens obtained from resection margins.

Comparison of qRT-PCR and ddPCR for multi-strain probiotic detection after a randomized human clinical trial.

The ability to detect probiotic consumption during a human clinical trial is crucial to verify and validate placebo and verum groups in post hoc analysis. While bacterial plating is still a common method for detecting and counting bacteria, when dealing with complex matrices like fecal samples, and given that most probiotics share genera or even species with commensal bacteria, plate counting is not a precise or accurate enough method. Species-specific quantitative real-time polymerase chain reaction (qRT-PCR) has been the most cited method in the literature and when properly validated and optimized remains the high watermark for detecting probiotics from fecal samples. Recent advancements in PCR technology have given rise to a parallel platform, droplet digital PCR (ddPCR). In this work we aimed to detect the components of a multi-strain probiotic product from a human clinical trial and compare both methods. This work dually demonstrates a process for determining multi-strain detection criteria as well as directly comparing the methods through the lens of sensitivity and specificity or the ability to properly discern true positives and true negatives. We described the optimization and validation of three assays for use in our detection panel and observed that, between qRT-PCR and ddPCR. The two methods were found to be quite congruent with ddPCR demonstrating a 10-100 fold lower limit of detection. Moreover, we discovered that most of the sensitivity and specificity had come from a single assay alone (Bifidobacterium animalis subsp. lactis Bl-04). This is despite all three assays performing well in optimization and validation. This suggests that more work needs to be done in the validation stage when developing novel probiotic detection assays. Taken together we can recommend ddPCR as a method for detecting probiotics from human clinical trials, but that qRT-PCR still performs well and comparably to ddPCR, when properly optimized and validated. However, when novel assays or those with unknown performance in a given biological matrix are needed, employing a strategy that combines multiple assays in a layered discrimination approach can help mitigate the potential underperformance of any single assay.

A Study on the Comparative Diagnostic Performance of Polymerase Chain Reaction, Rapid Diagnostic Technique and Microscopy in the three Northern Local Government Area of Taraba State, Nigeria

This study is conducted to compare diagnostic performance of Polymerase Chain Reaction, Rapid Diagnostic Techniqu`e and Microscopy. Nested PCR, RDT and Microscopy methods were used to screen for P. falciparum in the study populations. From site Jalingo, 85(75.9%), 70(62.5%) and 97(86.6%) samples were positive using nested PCR, RDT and Microscopy, while 27(24.1%), 42(37.5%) and 15(13.4%) were negative respectively. For site Zing, 85(73.3%) were positive for PCR, while 31(26.7%) were negative; 81(69.8) were positive for RDT and 35(30.1) were negative, while 104(89.7%) were positive with Microscopy and 12(10.3%) negative. Also, 82(75.2%), 50(45.9%) and 103(94.5%) were positive respectively for nested PCR, RDT and Microscopy, while 27(24.8%), 59(54.9%) and 6(5.5%) were negative respectively for the methods from site Lau. In all, microscopy method record the highest number of positive samples. The number of True Positive (TP) and True Negative (TN) recorded are 85 and 11 for PCR and 75 and 5 for Microscopy in Jalingo, 85 and 14 for PCR and 77 and 4 for Microscopy in Zing and 82 and 9 for PCR and 82 and 6 for Microscopy in Lau. The specificity and sensitivity of RDT from Jalingo, Zing and Lau are respectively 47.8% & 95.5%, 51.8% & 95.5% and 50.0% & 96.9%. Also, the specificity and sensitivity of Microscopy are 18.55% & 88.2%, 12.9% & 90.6% and 22.2% and 100% respectively. Owing to the higher sensitivity of the PCR method compared to Microscopy and RDT, P. falciparum detection by PCR was used as the reference method. From study site Jalingo, 22 (19.6%) and 5 (4.5%) false positive results Microscopy and RDT were negative for PCR and 10 (4.5%) and 20 (9.0%) false negatives Microscopy and RDT. This shows that Microscopy has a four-fold false positive detection error rate than RDT, while RDT has a two-fold false negative detection error rate. 56 (50%) were positive for the three methods. From study site Zing, the number of false positive for Microscopy and RDT are respectively 27 (23.3%) and 8 (6.9%), while PCR corrected that were false negatives were 8 (6.9%) and 12 (10.3%) respectively. Also from site Lau, 21 (19.3%) false positives were recorded for Microscopy, while only 2 (1.8%) were recorded for RDT. Microscopy had 0 (0%) false negatives while RDT has 34 (31.2%) that were nested PCR corrected. The same trend in the number of false positives and false negatives was observed in all the three sites, while Microscopy generally has higher false positive rate, RDT has a higher false negative rate. This study underscores the clinical utility of hematological and biochemical parameters in malaria management, particularly in resource-limited settings like Nigeria. The findings highlight the importance of comprehensive diagnostic approaches and suggest integrating these adjunct tools into malaria treatment protocols to enhance patient care and outcomes.

Diagnostic performance of a fully automated AI algorithm for lesion detection and PI-RADS classification in patients with suspected prostate cancer.

To evaluate the diagnostic performance of a fully automated, commercially available AIalgorithm for detecting prostate cancer and classifying lesions according to PI-RADS. In this retrospective single-center cohort study, we included consecutive patients with suspected prostate cancer who underwent 3T MRI between May 2017 and May 2020. Histopathological ground truth was targeted transperineal ultrasound-fusion guided biopsy and extensive systematic biopsy. We compared the results of the AI algorithm to those of human readers on both the lesion and patient level and determined the diagnostic performance. A total of 272 patients with 436 target lesions were evaluated. Of these patients, 135 (49.6%) had clinically significant prostate cancer (sPCa), 35 (12.9%) had clinically insignificant prostate cancer (ISUP = 1), and 102 (37.5%) were benign. On patient level, the cancer detection rates of sPCa for AI versus human readers were 11% versus 18% for PI-RADS ≤ 2, 27% versus 11% for PI-RADS 3, 54% versus 41% for PI-RADS 4, and 74% versus 92% for PI-RADS 5. The AI showed significantly higher accuracy: 74% versus 63% for PI-RADS ≥ 4 (p < 0.01) and 70% versus 52% for PI-RADS ≥ 3 (p < 0.01). Additionally, the AI correctly classified 62 patients with human reading PI-RADS ≥ 3 as true negatives. The AI algorithm proved to be a reliable and robust tool for lesion detection and classification. Its cancer detection rates and PI-RADS category distribution align with the results of recent meta-analyses, indicating precise risk stratification.

A Laboratory-Developed Assay for the Simultaneous Detection of Aspergillus fumigatus and Pneumocystis jirovecii Pulmonary Pathogens.

Invasive fungal diseases are a significant threat in immunocompromised patients, underscoring the need for rapid and accurate diagnostics. This study describes the development and validation of a real-time PCR-based laboratory-developed assay (LDA) on the Panther Fusion system for the simultaneous detection of Aspergillus fumigatus (AF) and Pneumocystis jirovecii (PJ) in bronchoalveolar lavage fluid (BALF) samples. The assay was evaluated using 239 clinical BALF samples, including cases confirmed positive for AF or PJ by reference mycological methods. Rigorous optimization ensured compatibility with the automated workflow of the Panther Fusion system, which addresses challenges such as BALF viscosity and fungal DNA recovery. No cross-reactivity with non-target fungal species was observed, and the assay demonstrated high analytical sensitivity and specificity. Only two false-negative results were reported, which could plausibly be reclassified as true negatives when interpreted alongside the serum beta-d-glucan and galactomannan assay results. For PJ detection, the assay showed excellent concordance with the OLM PneumID assay, supporting its reliability in clinical settings. The dual-target approach facilitates the simultaneous detection of both pathogens within a single workflow, improving diagnostic efficiency. The AF/PJ LDA represents a robust and scalable alternative to existing molecular assays, with the potential to enhance routine diagnostics for pulmonary fungal infections.

Analysis of Machine Learning Algorithms for Real-Time Gallbladder Stone Identification from Ultrasound Images in Clinical Decision Support Systems

Global epidemiology of gallstones in the twenty-first century affects millions of individuals, and ultrasound diagnostics effectively assess gallbladder size and function and detect abnormalities. This study collected datasets from local hospitals and reliable online sources for analysis using advanced CV/IP tools and WEKA. Image preprocessing techniques, including cropping, resizing, and grayscale conversion, were applied to 90 ultrasound images, extracting 600 ROIs with 21 features spanning binary, histogram, and texture attributes. The dataset was divided into balanced training and validation subsets, and supervised learning algorithms were optimized via cross-validation and grid search. Circular patterns were processed iteratively, with specific dimensions (512 × 512 for width/height, 32 × 32 for radius/blur, 128 × 128 for columns/rows). The performance of various machine learning classifiers was evaluated using accuracy, precision, recall, F1 score, AUC-ROC, MCC, Kappa GDR, and Dice Index, ensuring strong classification of normal and abnormal samples. The random forest (RF) classifier achieved the highest performance with an accuracy of 96.33%, followed by the MLP and Logit Boost classifiers with 95.67% and 95.40% accuracy rates, respectively. The RF model also exhibited the highest precision (0.9542), recall (0.9732), F1 score (0.9636), and a Dice Index (0.9649) with an MCC of 0.925, ROC area of 0.988, Kappa (0.921), and specificity of 95.34%-indicating its strong ability to balance true positives and negatives while minimizing misclassifications. The MLP classifier also performed well with a precision of 0.9477, a recall of 0.9665, and an F1 score of 0.957, while Logit Boost had similar results with a precision of 0.9411 and a recall of 0.9665. Other classifiers, such as the Bayes Net and J48 classifiers, showed slightly lower performance with accuracy rates of 94.67% but still exhibited good precision and recall, making them viable alternatives. This study highlights that the RF classifier achieved the highest superiority among other models in detecting gallbladder stones.

Diagnostic accuracy of artificial-intelligence-based electrocardiogram algorithm to estimate heart failure with reduced ejection fraction: A systematic review and meta-analysis.

Diagnostic accuracy of artificial-intelligence-based electrocardiogram algorithm to estimate heart failure with reduced ejection fraction: A systematic review and meta-analysis.

ROLE OF FINE NEEDLE ASPIRATION CYTOLOGY IN THE DIAGNOSIS OF THYROID DISEASE AT BOLAN MEDICAL COMPLEX HOSPITAL, QUETTA.

Background: Thyroid nodules are increasingly encountered in clinical practice, with a substantial proportion requiring diagnostic evaluation to exclude malignancy. Fine-needle aspiration cytology (FNAC) is widely utilized due to its minimally invasive nature, cost-effectiveness, and high diagnostic yield. Early and accurate differentiation between benign and malignant thyroid lesions is essential for appropriate surgical planning and to prevent overtreatment. This study investigates the diagnostic performance of FNAC in assessing thyroid nodules, using histopathology as the reference standard. Objective: To evaluate the diagnostic accuracy of fine-needle aspiration cytology (FNAC) in differentiating benign and malignant thyroid nodules. Methods: This cross-sectional study was conducted in the ENT Department of Bolan Medical Complex Hospital, Quetta, over six months, from November 2023 to May 2024. A total of 229 patients aged 18 to 75 years with clinically or radiologically suspected thyroid nodules underwent ultrasound-guided FNAC. The cytological findings were compared with postoperative histopathological outcomes. Inclusion criteria encompassed patients with suspected malignant thyroid nodules. Exclusion criteria included prior thyroid surgery, current treatment for thyroid malignancy, and non-consent. Data analysis was performed using IBM SPSS version 25.0. Diagnostic metrics including sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and overall accuracy were calculated. Results: Among 229 patients, 202 (88.21%) were female and 121 (52.84%) were between 18–45 years. FNAC results revealed 89 true positives, 10 false positives, 119 true negatives, and 11 false negatives. FNAC demonstrated a sensitivity of 89.0%, specificity of 92.25%, PPV of 89.90%, NPV of 91.54%, and an overall diagnostic accuracy of 90.83%. Conclusion: FNAC showed high diagnostic accuracy in distinguishing malignant from benign thyroid nodules, supporting its routine use in preoperative evaluation to guide surgical decision-making and improve patient outcomes.