Tool For Diagnosis Research Articles

Improving EEG Forward Modeling Using High-Resolution Five-Layer BEM-FMM Head Models: Effect on Source Reconstruction Accuracy

Electroencephalographic (EEG) source localization is a fundamental tool for clinical diagnoses and brain-computer interfaces. We investigate the impact of model complexity on reconstruction accuracy by comparing the widely used three-layer boundary element method (BEM) as an inverse method against a five-layer BEM accelerated by the fast multipole method (BEM-FMM) and coupled with adaptive mesh refinement (AMR) as forward solver. Modern BEM-FMM with AMR can solve high-resolution multi-tissue models efficiently and accurately. We generated noiseless 256-channel EEG data from 15 subjects in the Connectome Young Adult dataset, using four anatomically relevant dipole positions, three conductivity sets, and two head segmentations; we mapped localization errors across the entire grey matter from 4,000 dipole positions. The average location error among our four selected dipoles is ∼5mm (±2mm) with an orientation error of ∼12∘ (±7∘). The average source localization error across the entire grey matter is ∼9mm (±4mm), with a tendency for smaller errors on the occipital lobe. Our findings indicate that while three-layer models are robust under noiseless conditions, substantial localization errors (10–20mm) are common. Therefore, models of five or more layers may be needed for accurate source reconstruction in critical applications involving noisy EEG data.

Nutritional assessment in the HIV-infected older population receiving antiretroviral therapy.

Malnutrition is a pronounced public health issue which often seems underestimated in the older people living with HIV (PLWH) virus infection. PLWH are highly vulnerable to nutritional problems resulting from agingrelated deterioration, disease itself, and adverse effects of antiretroviral therapy (ART). The comprehensive nutritional assessments are necessary to perform routinely in this population to monitor and provide appropriate interventions to reduce comorbid conditions. In this review, we focus on the untoward impacts of malnutrition and nutritional assessments on the morbidity and mortality in the older PLWH. Some predictive factors of nutritional status in this group of patients are discussed. We propose the important components for nutrition assessment tool for older PLWH on ART. Highlighted issue is the need for developing uniform standardized tools for the early diagnosis of malnutrition in this population. Applications of the nutritional assessments, proper nutritional interventions, and regular monitoring of nutritional status in older PLWH liv-ing in every clinical setting may help the patients get better well-being.

Capture of Group A Streptococcus by Open-Microfluidic CandyCollect Device in Pediatric Patients.

Obtaining high-quality samples to diagnose streptococcal pharyngitis in pediatric patients is challenging due to discomfort associated with traditional pharyngeal swabs. This may cause reluctance to go to the clinic, inaccurate diagnosis, or inappropriate treatment for children with sore throat. Here, we determined the efficacy of CandyCollect, a lollipop-inspired open-microfluidic pathogen collection device, to capture Group A Streptococcus (GAS) and compare user preference for CandyCollect, conventional pharyngeal swabs, or mouth swabs in children with pharyngitis and their caregivers. All child participants (30/30) were positive for GAS by qPCR on both the mouth swab and CandyCollect. Caregivers ranked CandyCollect as a good sampling method overall (27/30), and all caregivers (30/30) would recommend CandyCollect for children 5 years and older. Twenty-three of 30 children "really like" the taste and 24/30 would prefer to use CandyCollect if a future test were needed. All caregivers (30/30) and most children (28/30) would be willing to use CandyCollect at home. All participants tested positive for GAS on all three collection methods (pharyngeal swab, mouth swab, and CandyCollect). While both caregivers and children like CandyCollect, some caregivers would prefer a shorter collection time. Future work includes additional studies with larger cohorts presenting with pharyngitis of unknown etiology and shortening collection time while maintaining the attractive form of the device. Obtaining oral samples for the diagnosis of streptococcal pharyngitis is of great importance for children. To address the challenges associated with traditional pharyngeal swab sampling, we developed the CandyCollect device, a lollipop-inspired open mesofluidic saliva sampling system. In this study, saliva samples were collected from children, aged 5-14 years, with CandyCollect and mouth swabs and analyzed via qPCR. The results show CandyCollect is the child preferred collection tool and had 100% concordance with the results from traditional diagnosis methods as part of their clinical care.

Mutations in ADAMTSL4 cause a unique form of autosomal-recessive congenital ectopia lentis

Several unique mutations of ADAMTSL4 leading to congenital ectopia lentis (CEL) have been previously reported by our team. The purpose of this study is to find out the possible mechanism of a recurrent novel intronic variant in ADAMTSL4 led to CEL. Twelve novel ADAMTSL4 mutations with a unique form congenital ectopic lentis were detected previously by panel-based NGS. Genetic analysis verified a novel heterozygous ADAMTSL4 variation c.2177+4A > G on Intron 11 in two unrelated patients with iris and lens abnormalities. MINI-Gene assay showed two splicing modes of mRNA that process two different bands A and B, and mutant-type shows replacement with the splicing mode of Exon 11 skipping. Construction of wild-type and mutant ADAMTSL4 vector showed the appearance of premature termination codons (PTC). In vitro expression detection showed significant down-regulated expression level of ADAMTSL4 mRNAs and proteins in cells transfected with mutant vectors compared with in wild-type group. On the contrary, translation inhibitor CHX and small interfering RNA of UPF1 (si-UPF1) significantly increased mRNA or protein expression of ADAMTSL4 in cells transfected with the mutant vectors. 12 novel mutations in ADAMTSL4 gene have been previously reported by our team in 6 CEL patients with a unique series of ocular abnormalities. The recurrent novel ADAMTSL4 mutation c.2177+4A > G triggering the splicing mode of Exon 11 skipping and NMD would cause the decrease of ADAMTSL4 proteins that participate in biosynthesis and assembly of microfibers, which might lead to CEL, and suggest that sequencing of certain intronic splicing varition might be a vital tool for genetic counseling and prenatal diagnoses.

Research on a Transformer Vibration Fault Diagnosis Method Based on Time-Shift Multiscale Increment Entropy and CatBoost.

A mechanical vibration fault diagnosis is a key means of ensuring the safe and stable operation of transformers. To achieve an accurate diagnosis of transformer vibration faults, this paper proposes a novel fault diagnosis method based on time-shift multiscale increment entropy (TSMIE) combined with CatBoost. Firstly, inspired by the concept of a time shift, TSMIE was proposed. TSMIE effectively solves the problem of the information loss caused by the coarse-graining process of traditional multiscale entropy. Secondly, the TSMIE of transformer vibration signals under different operating conditions was extracted as fault features. Finally, the features were sent into the CatBoost model for pattern recognition. Compared with different models, the simulation and experimental results showed that the proposed model had a higher diagnostic accuracy and stability, and this provides a new tool for transformer vibration fault diagnoses.

A generative adversarial network to Reinhard stain normalization for histopathology image analysis

Histopathology image analysis is paramount importance for accurate diagnosing diseases and gaining insight into tissue properties. The significant challenge of staining variability continues. This research work presents a new method that merges deep learning with Reinhardstain normalization, aiming to revolutionize histopathology image analysis. The multi-data stream attention-based generative adversarial network is an innovative architecture designed to enhance histopathological image analysis by integrating multiple data streams, attention mechanisms, and generative adversarial networks for improved feature extraction and image quality. Multi-data stream attention-based generative adversarial network capitalizes on attention mechanisms and generative adversarial networks to process multi-modal data efficiently, enhancing feature extraction and ensuring robust performance even in the presence of staining variations. This approach excels in exact disease detection and classification, emerging as an invaluable tool for both clinical diagnoses and research endeavors across diverse datasets. The obtained accuracy of the proposed method for the SCAN dataset is 97.75%, the BACH dataset is 99.50% and the Break His dataset is 99.66%. The proposed method significantly advances histopathology image analysis, offering improved diagnostic accuracy and deeper insights by integrating multi-data streams, attention mechanisms, and generative adversarial networks. This innovative approach enhances feature extraction, image quality, and overall effectiveness in medical image analysis.

Intelligent Reconfiguration-Promoted Cellular Internalization of Core-Shell DNA Nanoprobe Equipped with Successive Dual Stimuli-Responsive Protective Satellites for Amplification Fluorescence Imaging of Tumor Cells.

Although DNA probes have attracted increasing interest for precise tumor cell identification by imaging intracellular biomarkers, the requirement of commercial transfection reagents, limited targeting ligands, and/or non-biocompatible inorganic nanostructures has hampered the clinic translation. To circumvent these shortcomings, a reconfigurable ES-NC (Na+-dependent DNAzyme (E)-based substrate (S) cleavage core/shell DNA nanocluster (NC)) entirely from DNA strands is assembled for precise imaging of cancerous cells in a successive dual-stimuli-responsive manner. This nanoprobe is composed of a strung DNA tetrahedral satellites-based protective (DTP) shell, parallelly aligned target-responsive sensing (PTS) interlayer, and hydrophobic cholesterol-packed innermost layer (HCI core). Tetrahedral axial rotation-activated reconfiguration of DTP shell promotes the exposure of interior hydrophobic moieties, enabling cholesterol-mediated cellular internalization without auxiliary elements. Within cells, over-expressed glutathione triggers the disassembly of the DTP protective shell (first stimulus), facilitating target-stimulated signal transduction/amplification process (second stimuli). Target miRNA-21 is detected down to 10.6 fM without interference from coexisting miRNAs. Compared with transfection reagent-mediated counterpart, ES-NC displays a higher imaging ability, resists nuclease degradation, and has no detectable damage to healthy cells. The blindtest demonstrates that the ES-NC is suitable for the identification of cancerous cells from healthy cells, indicating a promising tool for early diagnosis and prediction of cancer.

Alzheimer’s disease classification algorithm based on fusion of channel attention and densely connected networks

Mild cognitive impairment (MCI) is a syndrome that occurs in the preclinical stage of Alzheimer’s disease (AD) and is also an early signal of the onset of AD. Early detection and accurate differentiation between MCI and AD populations, and providing them with effective intervention and treatment, are of great significance for preventing or delaying the onset of AD. In this paper, we propose a deep learning model, SE-DenseNet, that combines channel attention and dense connectivity networks and apply it to the field of magnetic resonance imaging (MRI) data recognition for the diagnosis of AD and MCI. First, to extract MRI features with high quality, a slicing algorithm based on two-dimensional image information entropy is proposed to obtain AD brain lesion features with stronger representation ability. Second, in terms of model structure, SENet is introduced as a channel attention module and redistribute the weight of image features in the channel dimension; use DenseNet as the main architecture to maximize information flow, and each layer is directly interconnected with subsequent layers. It enables the network to learn and extract relevant features from the input data and improve the classification ability of the network. Finally, our proposed model is validated on the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset, the results have shown that the accuracy for the four classification tasks of AD-NC, AD-MCI, NC-MCI, and AD-NC-MCI can reach 98.12%, 97.42%, 97.42%, and 95.24%, respectively. At the same time, the sensitivity and specificity have also achieved satisfactory results, exhibited a high performance in comparison with the classic machine learning algorithm and several existing state-of-the-art deep learning methods, demonstrating the proposed method is a powerful tool for the early diagnosis and detection of AD.

A Review on Graphene Analytical Sensors for Biomarker-based Detection of Cancer.

The engineering of nanoscale materials has broadened the scope of nanotechnology in a restricted functional system. Today, significant priority is given to immediate health diagnosis and monitoring tools for point-of-care testing and patient care. Graphene, as a one-atom carbon compound, has the potential to detect cancer biomarkers and its derivatives. The atom-wide graphene layer specialises in physicochemical characteristics, such as improved electrical and thermal conductivity, optical transparency, and increased chemical and mechanical strength, thus making it the best material for cancer biomarker detection. The outstanding mechanical, electrical, electrochemical, and optical properties of two-dimensional graphene can fulfil the scientific goal of any biosensor development, which is to develop a more compact and portable point-of-care device for quick and early cancer diagnosis. The bio-functionalisation of recognised biomarkers can be improved by oxygenated graphene layers and their composites. The significance of graphene that gleans its missing data for its high expertise to be evaluated, including the variety in surface modification and analytical reports. This review provides critical insights into graphene to inspire research that would address the current and remaining hurdles in cancer diagnosis.

A Learning-Based Approach for Diagnosis and Diagnosability of Unknown Discrete Event Systems.

This article develops a novel active-learning technique for fault diagnosis of an initially unknown finite-state discrete event system (DES). The proposed method constructs a diagnosis tool (termed diagnoser), which is able to detect and identify occurred faults by tracking the observable behaviors of the system under diagnosis. The proposed algorithm utilizes an active-learning mechanism to incrementally collect the information about the system to construct the diagnoser. This is achieved by completing a series of observation tables in a systematic way, resulting in the construction of the diagnoser. It is proven that the proposed algorithm terminates after a finite number of iterations and returns a correctly conjectured diagnoser. The developed diagnoser is a deterministic finite-state automaton. Furthermore, we have proven that the developed diagnoser consists of a minimum number of states. A sufficient condition for diagnosability of the system under diagnosis is derived, which guarantees the diagnosis of faults within a bounded number of observations. The developed method is applied to two case-studies, illustrating the steps of the proposed algorithm and its capability of diagnosing multiple faults.

Pneumonia Detection using CNN, Resnet and DenseNet

Abstract: Pneumonia is a prevalent respiratory infection that poses a significant threat to public health. Timely and precise diagnosis is essential for effective treatment and patient management. The proposed methodology involves training a CNN (Convolutional Neural Networks), ResNet-50 and DenseNet-121, on a large dataset of chest X ray images. CNNs play a crucial role in feature extraction from chest X-ray images for various computer vision tasks, including image recognition. The CNN model automatically learns and extracts essential visual features from the input images, capturing patterns and characteristics. On the other hand, ResNet-50 and DenseNet-121 leverage their effectiveness in handling deeper architectures and handling vanishing gradient problems. We compare our approach with existing methods to assess the quality and accuracy of the generated captions. The models undergo training and testing utilizing an extensive dataset comprising chest X-ray images, demonstrating high accuracy in detecting pneumonia, potentially offering a valuable tool for early diagnosis and treatment. The proposed pneumonia detection framework holds great promise for assisting healthcare professionals in diagnosing and treating pneumonia, thereby improving patient outcomes, and reducing healthcare costs.

Extended Enrichment for Ultrasensitive Detection of Low-Frequency Mutations by Long Blocker Displacement Amplification.

Detecting low-frequency DNA mutations hotspots cluster is critical for cancer diagnosis but remains challenging. Quantitative PCR (qPCR) is constrained by sensitivity, and allele-specific PCR is restricted by throughput. Here we develop a long blocker displacement amplification (LBDA) coupled with qPCR for ultrasensitive and multiplexed variants detection. By designing long blocker oligos to perfectly match wildtype sequences while mispairing with mutants, long blockers enable 14-44 nt enrichment regions which is 2-fold longer than normal BDA in the experiments. For wild template with a specific nucleotide, LBDA can detect different mutation types down to 0.5 % variant allele frequency (VAF) in one reaction, with median enrichment fold of 1,000 on 21 mutant DNA templates compared to the wild type. We applied LBDA-qPCR to detect KRAS and NRAS mutation hotspots, utilizing a single plex assay capable of covering 81 mutations and tested in synthetic templates and colorectal cancer tissue samples. Moreover, the mutation types were verified through Sanger sequencing, demonstrating concordance with results obtained from next generation sequencing. Overall, LBDA-qPCR provides a simple yet ultrasensitive approach for multiplexed detection of low VAF mutations hotspots, presenting a powerful tool for cancer diagnosis and monitoring.

Lossless compression-based detection of osteoporosis using bone X-ray imaging.

Digital X-ray imaging is essential for diagnosing osteoporosis, but distinguishing affected patients from healthy individuals using these images remains challenging. This study introduces a novel method using deep learning to improve osteoporosis diagnosis from bone X-ray images. A dataset of bone X-ray images was analyzed using a newly proposed procedure. This procedure involves segregating the images into regions of interest (ROI) and non-ROI, thereby reducing data redundancy. The images were then processed to enhance both spatial and statistical features. For classification, a Support Vector Machine (SVM) classifier was employed to distinguish between osteoporotic and non-osteoporotic cases. The proposed method demonstrated a promising Area under the Curve (AUC) of 90.8% in diagnosing osteoporosis, benchmarking favorably against existing techniques. This signifies a high level of accuracy in distinguishing osteoporosis patients from healthy controls. The proposed method effectively distinguishes between osteoporotic and non-osteoporotic cases using bone X-ray images. By enhancing image features and employing SVM classification, the technique offers a promising tool for efficient and accurate osteoporosis diagnosis.

Incidental pathogenic germline alterations detected through liquid biopsy in patients with solid tumors: prevalence, clinical utility and implications.

Liquid biopsy, a minimally invasive approach for detecting tumor biomarkers in blood, has emerged as a leading-edge technique in cancer precision medicine. New evidence has shown that liquid biopsies can incidentally detect pathogenic germline variants (PGVs) associated with cancer predisposition, including in patients with a cancer for which genetic testing is not recommended. The ability to detect these incidental PGV in cancer patients through liquid biopsy raises important questions regarding the management of this information and its clinical implications. This incidental identification of PGVs raises concerns about cancer predisposition and the potential impact on patient management, not only in terms of providing access to treatment based on the tumor molecular profiling, but also the management of revealing genetic predisposition in patients and families. Understanding how to interpret this information is essential to ensure proper decision-making and to optimize cancer treatment and prevention strategies. In this review we provide a comprehensive summary of current evidence of incidental PGVs in cancer predisposition genes identified by liquid biopsy in patients with cancer. We critically review the methodological considerations of liquid biopsy as a tool for germline diagnosis, clinical utility and potential implications for cancer prevention, treatment, and research.

Abstract 3448: A novel liquid biopsy lung cancer detection method using a coverage and fragment end motif machine learning analysis

Abstract Lung cancer remains the leading cause of cancer death in the United States. One reason for this high mortality is that >40% of cases are diagnosed at a late stage, where the 5-year survival rate is low. Liquid biopsy molecular testing has transformed lung cancer care through the identification of driver mutations when tissue is inaccessible. It also has promise as a tool in early detection and diagnosis, thereby reducing mortality. The existing paradigm for lung cancer screening and diagnosis includes low dose CT (LDCT) scans and tissue biopsy. LDCT scans yield a visualization of lung abnormalities. However, many of these abnormalities are difficult to characterize and next steps are often unclear to physicians. One option is to continue to monitor the patient, which may risk a delay in cancer treatment. Up to 15% of all individuals who are put on a monitoring regime will ultimately be diagnosed with lung cancer. A second option is to biopsy the nodule, which risks unnecessary invasive procedures and expenses to patients. 42-62% of indeterminate nodules detected by LDCT are found to be benign after biopsy. Furthermore, 15% of transthoracic needle biopsies result in complications. Genece Health aims to decrease uncertainty in cancer diagnosis and improve patient outcomes through a minimally invasive liquid biopsy assay that detects the presence of lung cancer. The technology utilizes a proprietary algorithm that leverages machine learning models to detect lung cancer signals in low-pass whole genome sequencing (LP-WGS) of patient cfDNA isolated from plasma. To establish assay performance, 50 clinical lung cancer samples from a commercial biobank, and 100 presumed normal samples were collected in Streck cfDNA BCT Devices. Following cfDNA extraction from plasma, LP-WGS was performed. Sequencing data was then processed to generate fragment end-motif and size (FEMS) and fragment coverage data that was input into Genece Health’s proprietary machine learning classifier to generate a final cancer prediction score. In addition, a limit of detection was determined using three late-stage lung cancer samples, with tumor fractions estimated using whole exome sequencing and ichorCNA scores. Our results show >80% specificity, >90% sensitivity, and a limit of detection of <1% tumor fraction. Through this testing, we demonstrate the assay’s ability to detect lung cancer across various lung cancer stages, including early stage disease, and histologies. Genece has developed a highly sensitive and specific lung cancer detection liquid biopsy assay. We present this assay as a cost-effective approach to discriminate between benign and malignant nodules identified by LDCT. In addition, this assay has the potential to be applied across the cancer care continuum. For example, future work will involve characterizing the assay’s ability to be applied in early cancer screening. Citation Format: Bryan Leatham, Michael Salmans, Molly Smith, Mengchi Wang, Andrew Carson, Kristin Fathe, Byung In Lee. A novel liquid biopsy lung cancer detection method using a coverage and fragment end motif machine learning analysis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3448.

Neurophysiology of hypokinetic movement’s disorders: New insights in daily clinical practice

Hypokinetic movement disorders encompass a group of clinically similar diseases that present challenges in discrimination during neurological examinations. Characterizing a specific hypokinetic disorder is necessary for the diagnosis and treatment in daily clinical practice. Neurophysiological tools, such as electromyography (EMG) combined with accelerometry, motor-evoked potentials (MEPs), electroencephalographic recording (EEG), Bereitschaftspotential (BP), auditory-evoked cognitive potential (P300), blink reflex (BR), and R2 blink reflex recovery cycle (R2BRRC), are useful in the differential diagnosis of movement disorders due to the common clinical features. However, neurophysiological assessments of movement disorders, especially hypokinetic diseases, are currently underutilized in clinical practice as compared to a few decades ago. This review aims to summarize practical insights gleaned from reported studies over the past 5 years (i.e., 2019 – 2023) regarding neurophysiological assessments of hypokinetic movement disorders, emphasizing the importance of their routine application. In particular, the methodology of the electrophysiologic evaluations pertaining to hypokinetic movement disorders is assessed. Moreover, a practical approach for the differential diagnosis of similar movement disorder syndromes based on specific neurophysiological features is proposed. Collectively, this review of the most recent neurophysiological implications in hypokinetic movement disorders highlights the practicality of these methods. Despite the advancement of other diagnostic techniques (e.g., neuroradiological methods), neurophysiological assessments may be a promising tool for clinical diagnoses, due to their high accuracy and ability to categorize and manage movement disorders (e.g., hypokinetic movement disorders) in daily clinical practice.

Comparative analysis of knee radiographs between weight bearing AP and Rosenberg PA flexion view for knee osteoarthrosis

: Osteoarthritis of the knee is highly prevalent in our country and a significant cause of morbidity. An early diagnosis is of paramount importance to provide treatment and delay it’s progression.: To evaluate the role of Rosenberg PA flexion view for early diagnosis of osteoarthritis, the degree of articular degeneration and the advantages of these views using Kellgren-Lawrence classification (KL), Ahlback grading (AB) and International Knee Documentation Committee (IKDC). Rosenberg PA flexion view significantly increased the detection of medial and lateral joint space narrowing (p<0.001) as compared to the standard weight bearing AP view alone. The severity of KL, Ahlback & IKDC scores increased significantly on Rosenberg view. The Rosenberg view increases the sensitivity of detection of JSN & the knee scores significantly & is an important tool for early diagnosis of OA.

Deep-learning-based super-resolution for accelerating chemical exchange saturation transfer MRI.

Chemical exchange saturation transfer (CEST) MRI is a molecular imaging tool that provides physiological information about tissues, making it an invaluable tool for disease diagnosis and guided treatment. Its clinical application requires the acquisition of high-resolution images capable of accurately identifying subtle regional changes in vivo, while simultaneously maintaining a high level of spectral resolution. However, the acquisition of such high-resolution images is time consuming, presenting a challenge for practical implementation in clinical settings. Among several techniques that have been explored to reduce the acquisition time in MRI, deep-learning-based super-resolution (DLSR) is a promising approach to address this problem due to its adaptability to any acquisition sequence and hardware. However, its translation to CEST MRI has been hindered by the lack of the large CEST datasets required for network development. Thus, we aim to develop a DLSR method, named DLSR-CEST, to reduce the acquisition time for CEST MRI by reconstructing high-resolution images from fast low-resolution acquisitions. This is achieved by first pretraining the DLSR-CEST on human brain T1w and T2w images to initialize the weights of the network and then training the network on very small human and mouse brain CEST datasetsto fine-tune the weights. Using the trained DLSR-CEST network, the reconstructed CEST source images exhibited improved spatial resolution in both peak signal-to-noise ratio and structural similarity index measure metrics at all downsampling factors (2-8). Moreover, amide CEST and relayed nuclear Overhauser effect maps extrapolated from the DLSR-CEST source images exhibited high spatial resolution and low normalized root mean square error, indicating a negligible loss in Z-spectrum information. Therefore, our DLSR-CEST demonstrated a robust reconstruction of high-resolution CEST source images from fast low-resolution acquisitions, thereby improving the spatial resolution and preserving most Z-spectrum information.

Extended Enrichment for Ultrasensitive Detection of Low‐Frequency Mutations by Long Blocker Displacement Amplification

AbstractDetecting low‐frequency DNA mutations hotspots cluster is critical for cancer diagnosis but remains challenging. Quantitative PCR (qPCR) is constrained by sensitivity, and allele‐specific PCR is restricted by throughput. Here we develop a long blocker displacement amplification (LBDA) coupled with qPCR for ultrasensitive and multiplexed variants detection. By designing long blocker oligos to perfectly match wildtype sequences while mispairing with mutants, long blockers enable 14–44 nt enrichment regions which is 2‐fold longer than normal BDA in the experiments. For wild template with a specific nucleotide, LBDA can detect different mutation types down to 0.5 % variant allele frequency (VAF) in one reaction, with median enrichment fold of 1,000 on 21 mutant DNA templates compared to the wild type. We applied LBDA‐qPCR to detect KRAS and NRAS mutation hotspots, utilizing a single plex assay capable of covering 81 mutations and tested in synthetic templates and colorectal cancer tissue samples. Moreover, the mutation types were verified through Sanger sequencing, demonstrating concordance with results obtained from next generation sequencing. Overall, LBDA‐qPCR provides a simple yet ultrasensitive approach for multiplexed detection of low VAF mutations hotspots, presenting a powerful tool for cancer diagnosis and monitoring.

Enhanced Artificial Intelligence Strategies in Renal Oncology: Iterative Optimization and Comparative Analysis of GPT 3.5 Versus 4.0.

The rise of artificial intelligence (AI) in medicine has revealed the potential of ChatGPT as a pivotal tool in medical diagnosis and treatment. This study assesses the efficacy of ChatGPT versions 3.5 and 4.0 in addressing renal cell carcinoma (RCC) clinical inquiries. Notably, fine-tuning and iterative optimization of the model corrected ChatGPT's limitations in this area. In our study, 80 RCC-related clinical questions from urology experts were posed three times to both ChatGPT 3.5 and ChatGPT 4.0, seeking binary (yes/no) responses. We then statistically analyzed the answers. Finally, we fine-tuned the GPT-3.5 Turbo model using these questions, and assessed its training outcomes. We found that the average accuracy rates of answers provided by ChatGPT versions 3.5 and 4.0 were 67.08% and 77.50%, respectively. ChatGPT 4.0 outperformed ChatGPT 3.5, with a higher accuracy rate in responses (p<0.05). By counting the number of correct responses to the 80 questions, we then found that although ChatGPT 4.0 performed better (p<0.05), both versions were subject to instability in answering. Finally, by fine-tuning the GPT-3.5 Turbo model, we found that the correct rate of responses to these questions could be stabilized at 93.75%. Iterative optimization of the model can result in 100% response accuracy. We compared ChatGPT versions 3.5 and 4.0 in addressing clinical RCC questions, identifying their limitations. By applying the GPT-3.5Turbo fine-tuned model iterative training method, we enhanced AI strategies in renal oncology. This approach is set to enhance ChatGPT's database and clinical guidance capabilities, optimizing AI in this field.