Low False Positive Rate Research Articles

DeepCTG® 1.0: an interpretable model to detect fetal hypoxia from cardiotocography data during labor and delivery.

Cardiotocography, which consists in monitoring the fetal heart rate as well as uterine activity, is widely used in clinical practice to assess fetal wellbeing during labor and delivery in order to detect fetal hypoxia and intervene before permanent damage to the fetus. We present DeepCTG® 1.0, a model able to predict fetal acidosis from the cardiotocography signals. DeepCTG® 1.0 is based on a logistic regression model fed with four features extracted from the last available 30 min segment of cardiotocography signals: the minimum and maximum values of the fetal heart rate baseline, and the area covered by accelerations and decelerations. Those four features have been selected among a larger set of 25 features. The model has been trained and evaluated on three datasets: the open CTU-UHB dataset, the SPaM dataset and a dataset built in hospital Beaujon (Clichy, France). Its performance has been compared with other published models and with nine obstetricians who have annotated the CTU-UHB cases. We have also evaluated the impact of two key factors on the performance of the model: the inclusion of cesareans in the datasets and the length of the cardiotocography segment used to compute the features fed to the model. The AUC of the model is 0.74 on the CTU-UHB and Beaujon datasets, and between 0.77 and 0.87 on the SPaM dataset. It achieves a much lower false positive rate (12% vs. 25%) than the most frequent annotation among the nine obstetricians for the same sensitivity (45%). The performance of the model is slightly lower on the cesarean cases only (AUC = 0.74 vs. 0.76) and feeding the model with shorter CTG segments leads to a significant decrease in its performance (AUC = 0.68 with 10 min segments). Although being relatively simple, DeepCTG® 1.0 reaches a good performance: it compares very favorably to clinical practice and performs slightly better than other published models based on similar approaches. It has the important characteristic of being interpretable, as the four features it is based on are known and understood by practitioners. The model could be improved further by integrating maternofetal clinical factors, using more advanced machine learning or deep learning approaches and having a more robust evaluation of the model based on a larger dataset with more pathological cases and covering more maternity centers.

Population size in QTL detection using quantile regression in genome-wide association studies

The aim of this study was to evaluate the performance of Quantile Regression (QR) in Genome-Wide Association Studies (GWAS) regarding the ability to detect QTLs (Quantitative Trait Locus) associated with phenotypic traits of interest, considering different population sizes. For this, simulated data was used, with traits of different levels of heritability (0.30 and 0.50), and controlled by 3 and 100 QTLs. Populations of 1,000 to 200 individuals were defined, with a random reduction of 100 individuals for each population. The power of detection of QTLs and the false positive rate were obtained by means of QR considering three different quantiles (0.10, 0.50 and 0.90) and also by means of the General Linear Model (GLM). In general, it was observed that the QR models showed greater power of detection of QTLs in all scenarios evaluated and a relatively low false positive rate in scenarios with a greater number of individuals. The models with the highest detection power of true QTLs at the extreme quantils (0.10 and 0.90) were the ones with the highest detection power of true QTLs. In contrast, the analysis based on the GLM detected few (scenarios with larger population size) or no QTLs in the evaluated scenarios. In the scenarios with low heritability, QR obtained a high detection power. Thus, it was verified that the use of QR in GWAS is effective, allowing the detection of QTLs associated with traits of interest even in scenarios with few genotyped and phenotyped individuals.

Rescuing biologically relevant consensus regions across replicated samples

BackgroundProtein-DNA binding sites of ChIP-seq experiments are identified where the binding affinity is significant based on a given threshold. The choice of the threshold is a trade-off between conservative region identification and discarding weak, but true binding sites.ResultsWe rescue weak binding sites using MSPC, which efficiently exploits replicates to lower the threshold required to identify a site while keeping a low false-positive rate, and we compare it to IDR, a widely used post-processing method for identifying highly reproducible peaks across replicates. We observe several master transcription regulators (e.g., SP1 and GATA3) and HDAC2-GATA1 regulatory networks on rescued regions in K562 cell line.ConclusionsWe argue the biological relevance of weak binding sites and the information they add when rescued by MSPC. An implementation of the proposed extended MSPC methodology and the scripts to reproduce the performed analysis are freely available at https://genometric.github.io/MSPC/; MSPC is distributed as a command-line application and an R package available from Bioconductor (https://doi.org/doi:10.18129/B9.bioc.rmspc).

Bayesian analysis with inaccurate rapid antigen testing for detecting someone is positive

This paper examines how testing accuracy plays a key role in detecting someone is positive. Bayesian-theorem exploits how an inaccurate rapid antigen test with a high true-positive rate and a low false-positive rate affects the detection of the probability that someone is positive. The probability is calculated with three determinants in actual dental practices: Infection rate, true positive rate, and false positive rate. The result suggests that in a high infection rate, the false positive rate of rapid antigen tests plays a key role in detecting positive individuals. In 6.5% infection rate, 0.991 true positive rate, and 0.0005 false positive rate, the probability is 0.992 while it with 0.05 false positive rate is reduced to 0.579. The proposed Bayesian analysis can be used for future analysis with imprecise tests in other applications.

Questionnaires for the Assessment of Cognitive Function Secondary to Intake Interviews in In-Hospital Work and Development and Evaluation of a Classification Model Using Acoustic Features.

The number of people with dementia is increasing each year, and early detection allows for early intervention and treatment. Since conventional screening methods are time-consuming and expensive, a simple and inexpensive screening is expected. We created a standardized intake questionnaire with thirty questions in five categories and used machine learning to categorize older adults with moderate and mild dementia and mild cognitive impairment, based on speech patterns. To evaluate the feasibility of the developed interview items and the accuracy of the classification model based on acoustic features, 29 participants (7 males and 22 females) aged 72 to 91 years were recruited with the approval of the University of Tokyo Hospital. The MMSE results showed that 12 participants had moderate dementia with MMSE scores of 20 or less, 8 participants had mild dementia with MMSE scores between 21 and 23, and 9 participants had MCI with MMSE scores between 24 and 27. As a result, Mel-spectrogram generally outperformed MFCC in terms of accuracy, precision, recall, and F1-score in all classification tasks. The multi-classification using Mel-spectrogram achieved the highest accuracy of 0.932, while the binary classification of moderate dementia and MCI group using MFCC achieved the lowest accuracy of 0.502. The FDR was generally low for all classification tasks, indicating a low rate of false positives. However, the FNR was relatively high in some cases, indicating a higher rate of false negatives.

Machine Learning in Cybersecurity: Techniques and Challenges

In the computer world, data science is the force behind the recent dramatic changes in cybersecurity's operations and technologies. The secret to making a security system automated and intelligent is to extract patterns or insights related to security incidents from cybersecurity data and construct appropriate data-driven models. Data science, also known as diverse scientific approaches, machine learning techniques, processes, and systems, is the study of actual occurrences via the use of data. Due to its distinctive qualities, such as flexibility, scalability, and the capability to quickly adapt to new and unknowable obstacles, machine learning techniques have been used in many scientific fields. Due to notable advancements in social networks, cloud and web technologies, online banking, mobile environments, smart grids, etc., cyber security is a rapidly expanding sector that requires a lot of attention. Such a broad range of computer security issues have been effectively addressed by various machine learning techniques. This article covers several machine-learning applications in cyber security. Phishing detection, network intrusion detection, keystroke dynamics authentication, cryptography, human interaction proofs, spam detection in social networks, smart meter energy consumption profiling, and security concerns with machine learning techniques themselves are all covered in this study. The methodology involves collecting a large dataset of phishing and legitimate instances, extracting relevant features such as email headers, content, and URLs, and training a machine-learning model using supervised learning algorithms. Machine learning models can effectively identify phishing emails and websites with high accuracy and low false positive rates. To enhance phishing detection, it is recommended to continuously update the training dataset to include new phishing techniques and to employ ensemble methods that combine multiple machine learning models for better performance.

XRecon: An Explainbale IoT Reconnaissance Attack Detection System Based on Ensemble Learning.

IoT devices have grown in popularity in recent years. Statistics show that the number of online IoT devices exceeded 35 billion in 2022. This rapid growth in adoption made these devices an obvious target for malicious actors. Attacks such as botnets and malware injection usually start with a phase of reconnaissance to gather information about the target IoT device before exploitation. In this paper, we introduce a machine-learning-based detection system for reconnaissance attacks based on an explainable ensemble model. Our proposed system aims to detect scanning and reconnaissance activity of IoT devices and counter these attacks at an early stage of the attack campaign. The proposed system is designed to be efficient and lightweight to operate in severely resource-constrained environments. When tested, the implementation of the proposed system delivered an accuracy of 99%. Furthermore, the proposed system showed low false positive and false negative rates at 0.6% and 0.05%, respectively, while maintaining high efficiency and low resource consumption.

Establishing and validating unbiased cut‐offs for tau PET with Gaussian mixture model in Alzheimer’s disease

AbstractBackgroundTau PET positivity (T+) is recommended in Alzheimer’s disease (AD) trials involving preclinical and prodromal AD. As unbiased cut‐offs are still lacking, we established and validated data‐driven T+ PET cut‐offs for AD.MethodData were from ADNI (estimation cohort) and Geneva memory center (validation cohort). 18F‐Flortaucipir standardized uptake value ratios (SUVr) were computed for a global temporal meta‐ROI and for cortical regions corresponding to Braak‐based stages I‐VI in amyloid negative (Aβ−) cognitively normal (CN), Aβ+ CN, Aβ+ MCI and AD. For Tau PET data‐driven cut‐off estimation, Gaussian mixture model (GMM) was applied on the global and regional SUVr distributions in Aβ− CN and AD (n = 269). The GMM cut‐offs’ performance was compared with published 18F‐Flortaucipir cut‐offs (Mattsson et al., 2017) by assessing the percentage of T+ subjects within each diagnostic group.ResultCut‐offs were 1.36 for temporal meta‐ROI and ranged 1.20‐1.39 for stages. In the validation cohort, all Aβ− CN were tau negative (T−), while T+ percentage increased along the Aβ+ continuum (CN<MCI<AD) for temporal meta‐ROI (33<64<75) and stages I‐II (50<73<75), III (17<55<58), IV (17<57<75), V (0<18<50) and VI (0<5<33). Previously published cut‐offs reported a higher frequency of T+ in the AD spectrum and identified T+ subjects in the Aβ‐ CN group.ConclusionGMMs unbiased 18F‐Flortaucipir SUVr cut‐offs showed lower T+ percentage in Aβ− CN than published cut‐offs, possibly indicating lower false positive rates and suggesting that they might be useful for the in‐vivo assessment of participants eligibility in future preclinical and prodromal AD trials.

DenseLens – Using DenseNet ensembles and information criteria for finding and rank-ordering strong gravitational lenses

ABSTRACT Convolutional neural networks (CNNs) are the state-of-the-art technique for identifying strong gravitational lenses. Although they are highly successful in recovering genuine lens systems with a high true-positive rate, the unbalanced nature of the data set (lens systems are rare), still leads to a high false positive rate. For these techniques to be successful in upcoming surveys (e.g. with Euclid) most emphasis should be set on reducing false positives, rather than on reducing false negatives. In this paper, we introduce densely connected neural networks (DenseNets) as the CNN architecture in a new pipeline-ensemble model containing an ensemble of classification CNNs and regression CNNs to classify and rank-order lenses, respectively. We show that DenseNets achieve comparable true positive rates but considerably lower false positive rates (when compared to residual networks; ResNets). Thus, we recommend DenseNets for future missions involving large data sets, such as Euclid, where low false positive rates play a key role in the automated follow-up and analysis of large numbers of strong gravitational lens candidates when human vetting is no longer feasible.

AI blood signature in common blood tests for detection of gastric cancer in a cohort of 190,000 individuals.

1500 Background: Gastric cancer is the fourth leading cause of cancer death worldwide. Lack of symptoms in the early stage of disease leads to delayed presentation and low overall survival rate. Early detection is key, but there is not yet an established test for non-invasive gastric cancer screening. Routine blood test panels, including complete blood count, liver function, renal function and clotting profiles, could be potentially useful in reflecting bodily processes related to cancer. Current literature and territory-wide statistical data analyses from Hong Kong have shown these subtle changes and differences in measured levels of subcomponents between individuals with and without gastric cancer, that are not recognized clinically, form recognizable patterns distinct to gastric cancer. In this study, we hypothesized that gastric cancer signatures could be identified in routine blood data. Deep learning AI algorithms are used to identify the gastric cancer signature. Results from testing and validation of the identified routine blood signature using a Big Clinical Data cohort are reported. Methods: This is a territory-wide healthcare database study using data from Hong Kong Hospital Authority data collaboration laboratory. All patients prescribed with medications for dyspepsia and with all blood tests components (CBC, LFT, RFT, clotting function) available in the period between 2004-2015 inclusive were included. In the period, data from 2004 to 09 and 2011 to 14 were used as training cohort and the rest were used as testing cohort. A blood gastric cancer signature was generated using deep learning AI algorithms, and applied to predict gastric cancer in Big Data population. Results: 193,117 patients were included in the captioned period, in whom 4,790 patients were diagnosed to have gastric cancer. 151,449 (3,815 pts with gastric cancer and 147,634 pts disease free) patients were included In training cohort. In the testing cohort in 2010 and 2015 (975 pts with gastric cancer and 40,693 pts disease-free), the tumor signature yielded a sensitivity of 0.79, specificity of 0.99, positive predictive value of 0.95 and a negative predictive value of 0.99. Conclusions: The hypothesis of the presence of a distinctive gastric cancer signature from parameters in routine blood test data is tested and validated in a large population cohort with dyspepsia. The signature has high accuracy and high sensitivity with low false positive and false negative rates. Multicenter prospective testing is being conducted in cohort indicated for OGD in Hong Kong.[Table: see text]

Anti-Phishing Tools: A Thorough Comparison of Features and Performance

Abstract: Phishing attacks are a common form of cyber-attacks that can result in data breaches, financial losses, and harm an organization's reputation. The purpose of the study is to evaluate and compare three popular anti-phishing tools' effectiveness in detecting and preventing such attacks. The research method involves conducting a study to identify best performing antiphishing tool. The findings show that some tools have high accuracy; others have a lower false positive rate. Overall, this research paper provides useful information about the effectiveness of anti-phishing tools and can help organizations select the most appropriate tool for their specific needs.

Real-Time Detection of Bud Degeneration in Oil Palms Using an Unmanned Aerial Vehicle

This paper presents a novel methodology for the early detection of oil palm bud degeneration based on computer vision. The proposed system uses the YOLO algorithm to detect diseased plants within the bud by analyzing images captured by a drone within the crop. Our system uses a drone equipped with a Jetson Nano embedded system to obtain complete images of crops with a 75% reduction in time and with 40% more accuracy compared to the traditional method. As a result, our system achieves a precision of 92% and a recall of 96%, indicating a high detection rate and a low false-positive rate. In real-time detection, the system is able to effectively detect diseased plants by monitoring an entire hectare of crops in 25 min. The system is also able to detect diseased plants other than those it was trained on with 43% precision. These results suggest that our methodology provides an effective and reliable means of early detection of bud degeneration in oil palm crops, which can prevent the spread of pests and improve crop production.

Comparison of four electrocardiographic criteria for the detection of cardiac abnormalities in Singapore athletes.

Sudden cardiac death in athletes is a rare occurrence, the most common cause being hypertrophic cardiomyopathy, which increases the risk of sustained ventricular tachycardia or ventricular fibrillation. Most of these young athletes are asymptomatic prior to the cardiac arrest. Several electrocardiogram criteria such as the European Society of Cardiology group 2 Criteria changes, Seattle Criteria, Refined Criteria, and most recently the 2017 International Criteria, have sought to improve the accuracy of identifying these at-risk athletes during pre-participation screening while minimising unnecessary investigations for the majority of athletes at low risk.We aimed to compare the above four criteria in our Singapore athlete population to identify which criterion performed the best in detecting cardiac abnormalities on echocardiography. Out of 1,515 athletes included in Changi General Hospital, Singapore registry between June 2007 and June 2014, the electrocardiograms of 270 athletes with further cardiac investigations were analysed. We compared the above four electrocardiographic criteria to evaluate which performed best for detecting cardiac abnormalities on echocardiography in our Southeast Asian athlete population. The European Society of Cardiology, Seattle, Refined and 2017 International Criteria had a sensitivity of 20%, 0%, 20% and 5%, respectively; a specificity of 64%, 93%, 84% and 97%, respectively; a positive predictive value of 4%, 0%, 9% and 11%, respectively; and a negative predictive value of 91%, 92%, 93% and 93%, respectively for detecting abnormalities on echocardiography. The latest 2017 International Criteria performed the best as it had the highest specificity and positive predictive value, joint highest negative predictive value, and lowest false positive rate.

MDAP: Module Dependency based Anomaly Prediction

In large-scale distributed systems with multiple interconnected modules, failure of even a single module might have a cascading effect and might result in overall system failure. Naturally, timely identification and resolution of these system-wide anomaly is highly crucial for ensuring robust functioning of a system. However, while many frameworks for detecting anomalous behaviour exist today, very few frameworks are proposed which can leverage those anomalies for ultimately detecting system failures (note that some anomalies might not lead to a failure too). To fill this gap, in our work, we propose MDAP, a system anomaly-prediction framework for large-scale distributed systems. Specifically MDAP made the following contributions: (a) It detects anomalous behaviour in a system by considering behavioural changes of the interdependencies across different modules of the system with only information from system logs. (b) Provides the ability to raise early failure alerts. (c) Finally evaluating with three different datasets, MDAP achieves 39%–52% better accuracy compared to the baseline algorithms in identifying periods when the system is performing abnormally—we achieve a high true positive rate of 70.78% and low false positive rate of 3.56% using information about only a few bugs.

Securing IoT Devices Running PureOS from Ransomware Attacks: Leveraging Hybrid Machine Learning Techniques

Internet-enabled (IoT) devices are typically small, low-powered devices used for sensing and computing that enable remote monitoring and control of various environments through the Internet. Despite their usefulness in achieving a more connected cyber-physical world, these devices are vulnerable to ransomware attacks due to their limited resources and connectivity. To combat these threats, machine learning (ML) can be leveraged to identify and prevent ransomware attacks on IoT devices before they can cause significant damage. In this research paper, we explore the use of ML techniques to enhance ransomware defense in IoT devices running on the PureOS operating system. We have developed a ransomware detection framework using machine learning, which combines the XGBoost and ElasticNet algorithms in a hybrid approach. The design and implementation of our framework are based on the evaluation of various existing machine learning techniques. Our approach was tested using a dataset of real-world ransomware attacks on IoT devices and achieved high accuracy (90%) and low false-positive rates, demonstrating its effectiveness in detecting and preventing ransomware attacks on IoT devices running PureOS.

Detection and quantification of cotton trichomes by deep learning algorithm

Cotton is one of the most economically important agricultural crops with the harvest of cotton fibers. Trichomes are initiated from the plant epidermis, which is responsible for plant resistance to many abiotic and biotic stresses. Cotton fibers share similar underlying developmental molecular mechanisms with trichomes on the stem and leaf margins of cotton. Compared with fibers, it is easier to observe the trichome phenotype of cotton stem and leaf margins. However, because of the variability in coloration and the high occlusion of the trichomes, counting the density and observing the distribution of trichomes is still highly laborious. To help resolve this problem, the present research provides a solution based on deep learning in the screening of cotton trichomes, which makes it possible to monitor the phenotype of the cotton efficiently. With a fluorescent microscopy imaging procedure, we clearly observed that there are two types of trichome phenotype inGossypium hirsutum(‘TM-1′), which are singled and clustered trichomes. The clustered trichome originates from two or more adjacent epidermal cells with no branches. The TM1 trichome was used to create a collection of singled and clustered trichome pictures. Following data production, we evaluated the effectiveness of several you only look once (YOLO) model iterations in the context of identifying singled and clustered trichomes, respectively. Results show that the YOLOv5s model outperforms YOLOv3 and YOLOv4 in average accuracy, recall, accuracy, and F1 score. We utilized the clustered trichome data set to assess the performance of YOLOv5s and Mask R-CNN in order to more precisely detect the real trichome distribution with simultaneous detection of single and clustered trichomes. The obtained findings show that both models may be utilized to concurrently identify singled and clustered trichomes since the discrepancy between the detected values of the two models and the real values is quite minimal. Mask R-CNN has a high IOU value, whereas YOLOv5s has a low false-positive detection rate and fast detection speed. Our models and datasets can be used as a starting point to measure the trichome trait in other cotton species and establish a cotton trichome detection and counting system based on deep learning technology.

SEPPA-mAb: spatial epitope prediction of protein antigens for mAbs.

Identifying the exact epitope positions for a monoclonal antibody (mAb) is of critical importance yet highly challenging to the Ab design of biomedical research. Based on previous versions of SEPPA 3.0, we present SEPPA-mAb for the above purpose with high accuracy and low false positive rate (FPR), suitable for both experimental and modelled structures. In practice, SEPPA-mAb appended a fingerprints-based patch model to SEPPA 3.0, considering the structural and physic-chemical complementarity between a possible epitope patch and the complementarity-determining region of mAb and trained on 860 representative antigen-antibody complexes. On independent testing of 193 antigen-antibody pairs, SEPPA-mAb achieved an accuracy of 0.873 with an FPR of 0.097 in classifying epitope and non-epitope residues under the default threshold, while docking-based methods gave the best AUC of 0.691, and the top epitope prediction tool gave AUC of 0.730 with balanced accuracy of 0.635. A study on 36 independent HIV glycoproteins displayed a high accuracy of 0.918 and a low FPR of 0.058. Further testing illustrated outstanding robustness on new antigens and modelled antibodies. Being the first online tool predicting mAb-specific epitopes, SEPPA-mAb may help to discover new epitopes and design better mAbs for therapeutic and diagnostic purposes. SEPPA-mAb can be accessed at http://www.badd-cao.net/seppa-mab/.

Sensitivity and specificity of conventional and new face validation in determining the incomprehensible items by older people: Empirical evidence of testing 106 quality-of-life items.

This methodological research aimed to investigate and compare the sensitivity and specificity of conventional and new face validation in identifying incomprehensible items empirically. A purposive sample of 15 older people living in three residential care homes (RCHs) in Hong Kong was used to evaluate a newly developed 106 items covering seven quality-of-life dimensions. The abbreviated Mental Test (Hong Kong version; AMT) was used as a screening tool for excluding those with impaired cognition. The interview was audiotaped, and incomprehensible items were identified by the research panel accordingly (served as the gold standard). The socio-demographics of the respondents were described. Understandability (yes/no, conventional face validation method) and interpretability (4-point Likert scale, new method) were compared and used to compute the Kappa value (representing chance agreement), sensitivity, and specificity analysis. Fifteen older people were interviewed and responded to the structured interview of 106 items regarding understandability and interpretability. 61 items (57%) obtained 100% positive understandability while only 35 items (33%) obtained 100% correct interpretability.The Kappa coefficient was 0.388 (P < 0.001) of the chance agreement between understandability and interpretability. The panel confirmed that 32% of items required revision (i.e., incomprehensible items). The false negative rate of using the conventional approach was up to 70.59% while both the false positive and negative rates of using the new approach were low (0%-5.88%). This empirical evidence indicated that the conventional approach of face validation for checking incomprehensible items by older people encountered a high false negative rate. On the contrary, the new approach was recommended because it demonstrated high sensitivity and specificity and low false positive and negative rates in identifying incomprehensible items.

A hybrid learning method for distinguishing lung adenocarcinoma and squamous cell carcinoma

PurposeThe objective of the proposed work is to identify the most commonly occurring non–small cell carcinoma types, such as adenocarcinoma and squamous cell carcinoma, within the human population. Another objective of the work is to reduce the false positive rate during the classification.Design/methodology/approachIn this work, a hybrid method using convolutional neural networks (CNNs), extreme gradient boosting (XGBoost) and long-short-term memory networks (LSTMs) has been proposed to distinguish between lung adenocarcinoma and squamous cell carcinoma. To extract features from non–small cell lung carcinoma images, a three-layer convolution and three-layer max-pooling-based CNN is used. A few important features have been selected from the extracted features using the XGBoost algorithm as the optimal feature. Finally, LSTM has been used for the classification of carcinoma types. The accuracy of the proposed method is 99.57 per cent, and the false positive rate is 0.427 per cent.FindingsThe proposed CNN–XGBoost–LSTM hybrid method has significantly improved the results in distinguishing between adenocarcinoma and squamous cell carcinoma. The importance of the method can be outlined as follows: It has a very low false positive rate of 0.427 per cent. It has very high accuracy, i.e. 99.57 per cent. CNN-based features are providing accurate results in classifying lung carcinoma. It has the potential to serve as an assisting aid for doctors.Practical implicationsIt can be used by doctors as a secondary tool for the analysis of non–small cell lung cancers.Social implicationsIt can help rural doctors by sending the patients to specialized doctors for more analysis of lung cancer.Originality/valueIn this work, a hybrid method using CNN, XGBoost and LSTM has been proposed to distinguish between lung adenocarcinoma and squamous cell carcinoma. A three-layer convolution and three-layer max-pooling-based CNN is used to extract features from the non–small cell lung carcinoma images. A few important features have been selected from the extracted features using the XGBoost algorithm as the optimal feature. Finally, LSTM has been used for the classification of carcinoma types.

Phylogeny-based assignment of functional traits to DNA barcodes outperforms distance-based, in a comparison of approaches.

The full potential for using DNA barcodes for profiling functional trait diversity has yet to be determined in plants and animals; thus, we outline a general framework for quantifying functional trait diversity of insect community DNA and propose and assess the accuracy of three methods for achieving this. We built a novel dataset of traits and DNA barcodes for wild bees in China. An informatics framework was developed for phylogeny-based integration of these data and prediction of traits for any subject barcodes, which was compared with two distance-based methods. For Phylogenetic Assignment, we additionally conducted a species-level analysis of publically available bee trait data. Under the specimen-level dataset, the rate of trait assignment was negatively correlated with distance between the query and the nearest trait-known reference, for all methods. Phylogenetic Assignment was found to perform best under several criteria; particularly, it had the lowest false-positive rate (rarely returning a state prediction where success was unlikely; where the distance from query to the nearest reference was high). For a wider range of compiled traits, conservative life-history traits showed the highest rates of assignment; for example, sociality was predicted with confidence at 53%, parasitism at 44% and nest location at 33%. As outlined herein, automated trait assignment might be applied at scale to either barcodes or metabarcodes. With further compilation and databasing of DNA barcode and trait data, the rate and accuracy of trait assignment is expected to increase to the point of being a widely viable and informative approach.