Integration of temporal & spatial properties of dynamic functional connectivity based on two-directional two-dimensional principal component analysis for disease analysis.

Dynamic functional connectivity, derived from resting-state functional magnetic resonance imaging (rs-fMRI), has emerged as a crucial instrument for investigating and supporting the diagnosis of neurological disorders. However, prevalent features of dynamic functional connectivity predominantly capture either temporal or spatial properties, such as mean and global efficiency, neglecting the significant information embedded in the fusion of spatial and temporal attributes. In addition, dynamic functional connectivity suffers from the problem of temporal mismatch, i.e., the functional connectivity of different subjects at the same time point cannot be matched. To address these problems, this article introduces a novel feature extraction framework grounded in two-directional two-dimensional principal component analysis. This framework is designed to extract features that integrate both spatial and temporal properties of dynamic functional connectivity. Additionally, we propose to use Fourier transform to extract temporal-invariance properties contained in dynamic functional connectivity. Experimental findings underscore the superior performance of features extracted by this framework in classification experiments compared to features capturing individual properties.

Insights into early recovery from Long COVID-results from the German DigiHero Cohort.

65 million people worldwide are estimated to suffer from long-term symptoms after their SARS-CoV-2 infection (Long COVID). However, there is still little information about the early recovery among those who initially developed Long COVID, i.e. had symptoms 4-12 weeks after infection but no symptoms after 12 weeks. We aimed to identify associated factors with this early recovery. We used data from SARS-CoV-2-infected individuals from the DigiHero study. Participants provided information about their SARS-CoV-2 infections and symptoms at the time of infection, 4-12 weeks, and more than 12 weeks post-infection. We performed multivariable logistic regression to identify factors associated with early recovery from Long COVID and principal component analysis (PCA) to identify groups among symptoms. 5098 participants reported symptoms at 4-12 weeks after their SARS-CoV-2 infection, of which 2441 (48%) reported no symptoms after 12 weeks. Men, younger participants, individuals with mild course of acute infection, individuals infected with the Omicron variant, and individuals who did not seek medical care in the 4-12 week period after infection had a higher chance of early recovery. In the PCA, we identified four distinct symptom groups. Our results indicate differential risk of continuing symptoms among individuals who developed Long COVID. The identified risk factors are similar to those for the development of Long COVID, so people with these characteristics are at higher risk not only for developing Long COVID, but also for longer persistence of symptoms. Those who sought medical help were also more likely to have persistent symptoms.

Standardised Versioning of Datasets: a FAIR-compliant Proposal.

This paper presents a standardised dataset versioning framework for improved reusability, recognition and data version tracking, facilitating comparisons and informed decision-making for data usability and workflow integration. The framework adopts a software engineering-like data versioning nomenclature ("major.minor.patch") and incorporates data schema principles to promote reproducibility and collaboration. To quantify changes in statistical properties over time, the concept of data drift metrics (d) is introduced. Three metrics (dP, dE,PCA, and dE,AE) based on unsupervised Machine Learning techniques (Principal Component Analysis and Autoencoders) are evaluated for dataset creation, update, and deletion. The optimal choice is the dE,PCA metric, combining PCA models with splines. It exhibits efficient computational time, with values below 50 for new dataset batches and values consistent with seasonal or trend variations. Major updates (i.e., values of 100) occur when scaling transformations are applied to over 30% of variables while efficiently handling information loss, yielding values close to 0. This metric achieved a favourable trade-off between interpretability, robustness against information loss, and computation time.

Detection of lung cancer through SERS analysis of serum.

Early detection and postoperative assessment are crucial for improving overall survival among lung cancer patients. Here, we report a non-invasive technique that integrates Raman spectroscopy with machine learning for the detection of lung cancer. The study encompassed 88 postoperative lung cancer patients, 73 non-surgical lung cancer patients, and 68 healthy subjects. The primary aim was to explore variations in serum metabolism across these cohorts. Comparative analysis of average Raman spectra was conducted, while principal component analysis was employed for data visualization. Subsequently, the augmented dataset was used to train convolutional neural networks (CNN) and Resnet models, leading to the development of a diagnostic framework. The CNN model exhibited superior performance, as verified by the receiver operating characteristic curve. Notably, postoperative patients demonstrated an increased likelihood of recurrence, emphasizing the crucial need for continuous postoperative monitoring. In summary, the integration of Raman spectroscopy with CNN-based classification shows potential for early detection and postoperative assessment of lung cancer.

Clusterization of Behavioral and Psychological Symptoms of Dementia as Assessed by Neuropsychiatric Inventory: A Case Against the Use of Principal Component Analysis.

Background: The term Behavioral and Psychological Symptoms of Dementia (BPSD) covers a group of phenomenologically and medically distinct symptoms that rarely occur in isolation. Their therapy represents a major unmet medical need across dementias of different types, including Alzheimer's disease. Understanding of the symptom occurrence and their clusterization can inform clinical drug development and use of existing and future BPSD treatments. Objective: The primary aim of the present study was to investigate the ability of a commonly used principal component analysis to identify BPSD patterns as assessed by Neuropsychiatric Inventory (NPI). Methods: NPI scores from the Aging, Demographics, and Memory Study (ADAMS) were used to characterize reported occurrence of individual symptoms and their combinations. Based on this information, we have designed and conducted a simulation experiment to compare Principal Component analysis (PCA) and zero-inflated PCA (ZI PCA) by their ability to reveal true symptom associations. Results: Exploratory analysis of the ADAMS database revealed overlapping multivariate distributions of NPI symptom scores. Simulation experiments have indicated that PCA and ZI PCA cannot handle data with multiple overlapping patterns. Although the principal component analysis approach is commonly applied to NPI scores, it is at risk to reveal BPSD clusters that are a statistical phenomenon rather than symptom associations occurring in clinical practice. Conclusions: We recommend the thorough characterization of multivariate distributions before subjecting any dataset to Principal Component Analysis.

Egg vs. Oil in the Cookbook of Plasters: Differentiation of Lipid Binders in Wall Paintings Using Gas Chromatography-Mass Spectrometry and Principal Component Analysis.

Wall paintings are integral to cultural heritage and offer rich insights into historical and religious beliefs. There exist various wall painting techniques that pose challenges in binder and pigment identification, especially in the case of egg/oil-based binders. GC-MS identification of lipidic binders relies routinely on parameters like the ratios of fatty acids within the plaster. However, the reliability of these ratios for binder identification is severely limited, as demonstrated in this manuscript. Therefore, a more reliable tool for effective differentiation between egg and oil binders based on a combination of diagnostic values, specific markers (cholesterol oxidation products), and PCA is presented in this study. Reference samples of wall paintings with egg and linseed oil binders with six different pigments were subjected to modern artificial ageing methods and subsequently analysed using two GC-MS instruments. A statistically significant difference (at a 95% confidence level) between the egg and oil binders and between the results from two GC-MS instruments was observed. These discrepancies between the results from the two GC-MS instruments are likely attributed to the heterogeneity of the samples with egg and oil binders. This study highlights the complexities in identifying wall painting binders and the need for innovative and revised analytical methods in conservation efforts.

Low-abundance proteins-based label-free SERS approach for high precision detection of liver cancer with different stages.

BACKGROUND: Surface-enhanced Raman scattering (SERS) technology have unique advantages of rapid, simple, and highly sensitive in the detection of serum, it can be used for the detection of liver cancer. However, some protein biomarkers in body fluids are often present at ultra-low concentrations and severely interfered with by the high-abundance proteins (HAPs), which will affect the detection of specificity and accuracy in cancer screening based on the SERS immunoassay. Clearly, there is a need for an unlabeled SERS method based on low abundance proteins, which is rapid, noninvasive, and capable of high precision detection and screening of liver cancer. RESULTS: Serum samples were collected from 60 patients with liver cancer (27 patients with stage T1 and T2 liver cancer, 33 patients with stage T3 and T4 liver cancer) and 40 healthy volunteers. Herein, immunoglobulin and albumin were separated by immune sorption and Cohn ethanol fractionation. Then, the low abundance protein (LAPs) was enriched, and high-quality SERS spectral signals were detected and obtained. Finally, combined with the principal component analysis-linear discriminant analysis (PCA-LDA) algorithm, the SERS spectrum of early liver cancer (T1-T2) and advanced liver cancer (T3-T4) could be well distinguished from normal people, and the accuracy rate was 98.5% and 100%, respectively. Moreover, SERS technology based on serum LAPs extraction combined with the partial least square-support vector machine (PLS-SVM) successfully realized the classification and prediction of normal volunteers and liver cancer patients with different tumor (T) stages, and the diagnostic accuracy of PLS-SVM reached 87.5% in the unknown testing set. SIGNIFICANCE: The experimental results show that the serum LAPs SERS detection combined with multivariate statistical algorithms can be used for effectively distinguishing liver cancer patients from healthy volunteers, and even achieved the screening of early liver cancer with high accuracy (T1 and T2 stage). These results showed that serum LAPs SERS detection combined with a multivariate statistical diagnostic algorithm has certain application potential in early cancer screening.

Sensorimotor mechanisms selective to numerosity derived from individual differences.

We have previously shown that after few seconds of adaptation by finger-tapping, the perceived numerosity of spatial arrays and temporal sequences of visual objects displayed near the tapping region is increased or decreased, implying the existence of a sensorimotor numerosity system (Anobile et al., 2016). To date, this mechanism has been evidenced only by adaptation. Here, we extend our finding by leveraging on a well-established covariance technique, used to unveil and characterize 'channels' for basic visual features such as colour, motion, contrast, and spatial frequency. Participants were required to press rapidly a key a specific number of times, without counting. We then correlated the precision of reproduction for various target number presses between participants. The results showed high positive correlations for nearby target numbers, scaling down with numerical distance, implying tuning selectivity. Factor analysis identified two factors, one for low and the other for higher numbers. Principal component analysis revealed two bell-shaped covariance channels, peaking at different numerical values. Two control experiments ruled out the role of non-numerical strategies based on tapping frequency and response duration. These results reinforce our previous reports based on adaptation, and further suggest the existence of at least two sensorimotor number channels responsible for translating symbolic numbers into action sequences.

Deficits in force production during multifinger tasks demarcate cognitive dysfunction.

BACKGROUND: The multifinger force deficit (MFFD) is the decline in force generated by each finger as the number of fingers contributing to an action is increased. It has been shown to associate with cognitive status. AIMS: The aim was to establish whether a particularly challenging form of multifinger grip dynamometry, that provides minimal tactile feedback via cutaneous receptors and requires active compensation for reaction forces, will yield an MFFD that is more sensitive to cognitive status. METHODS: Associations between measures of motor function, and cognitive status (Montreal Cognitive Assessment [MoCA]) and latent components of cognitive function (derived from 11 tests using principal component analysis), were estimated cross-sectionally using generalized partial rank correlations. The participants (n = 62) were community dwelling, aged 65-87. RESULTS: Approximately half the participants were unable to complete the dynamometry task successfully. Cognitive status demarcated individuals who could perform the task from those who could not. Among those who complied with the task requirements, the MFFD was negatively correlated with MoCA scores-those with the highest MoCA scores tended to exhibit the smallest deficits, and vice versa. There were corresponding associations with latent components of cognitive function. DISCUSSION: The results support the view that neurodegenerative processes that are a feature of normal and pathological aging exert corresponding effects on expressions of motor coordination-in multifinger tasks, and cognitive sufficiency, due to their dependence on shared neural systems. CONCLUSIONS: The outcomes add weight to the assertion that deficits in force production during multifinger tasks are sensitive to cognitive dysfunction.

Principal Component Analysis Enhanced with Bootstrapped Confidence Interval for the Classification of Parkinsonian Patients Using Gaussian Mixture Model and Gait Initiation Parameters.

Parkinson's disease is one of the major neurodegenerative diseases that affects the postural stability of patients, especially during gait initiation. There is actually an increasing demand for the development of new non-pharmacological tools that can easily classify healthy/affected patients as well as the degree of evolution of the disease. The experimental characterization of gait initiation (GI) is usually done through the simultaneous acquisition of about 20 variables, resulting in very large datasets. Dimension reduction tools are therefore suitable, considering the complexity of the physiological processes involved. The principal Component Analysis (PCA) is very powerful at reducing the dimensionality of large datasets and emphasizing correlations between variables. In this paper, the Principal Component Analysis (PCA) was enhanced with bootstrapping and applied to the study of the GI to identify the 3 majors sets of variables influencing the postural control disability of Parkinsonian patients during GI. We show that the combination of these methods can lead to a significant improvement in the unsupervised classification of healthy/affected patients using a Gaussian mixture model, since it leads to a reduced confidence interval on the estimated parameters. The benefits of this method for the identification and study of the efficiency of potential treatments is not addressed in this paper but could be addressed in future works.

Embodimetrics: A Principal Component Analysis Study of the Combined Assessment of Cardiac, Cognitive and Mobility Parameters.

There is a growing body of literature investigating the relationship between the frequency domain analysis of heart rate variability (HRV) and cognitive Stroop task performance. We proposed a combined assessment integrating trunk mobility in 72 healthy women to investigate the relationship between cognitive, cardiac, and motor variables using principal component analysis (PCA). Additionally, we assessed changes in the relationships among these variables after a two-month intervention aimed at improving the perception-action link. At baseline, PCA correctly identified three components: one related to cardiac variables, one to trunk motion, and one to Stroop task performance. After the intervention, only two components were found, with trunk symmetry and range of motion, accuracy, time to complete the Stroop task, and low-frequency heart rate variability aggregated into a single component using PCA. Artificial neural network analysis confirmed the effects of both HRV and motor behavior on cognitive Stroop task performance. This analysis suggested that this protocol was effective in investigating embodied cognition, and we defined this approach as "embodimetrics".

In situ rapid evaluation method of quality of peach kernels based on near infrared spectroscopy.

This study aimed to perform a rapid in situ assessment of the quality of peach kernels using near infrared (NIR) spectroscopy, which included identifications of authenticity, species, and origins, and amygdalin quantitation. The in situ samples without any pretreatment were scanned by a portable MicroNIR spectrometer, while their powder samples were scanned by a benchtop Fourier transform NIR (FT-NIR) spectrometer. To improve the performance of the in situ determination model of the portable NIR spectrometer, the two spectrometers were first compared in identification and content models of peach kernels for both in situ and powder samples. Then, the in situ sample spectra were transferred by using the improved principal component analysis (IPCA) method to enhance the performance of the in situ model. After model transfer, the prediction performance of the in situ sample model was significantly improved, as shown by the correlation coefficient in the prediction set (Rp), root means square error of prediction (RMSEP), and residual prediction deviation (RPD) of the in situ model reached 0.9533, 0.0911, and 3.23, respectively, and correlation coefficient in the test set (Rt) and root means square error of test (RMSET) reached 0.9701 and 0.1619, respectively, suggesting that model transfer could be a viable solution to improve the model performance of portable spectrometers.

[Ecological Environment Assessment and Driving Mechanism Analysis of Nagqu and Amdo Sections of Qinghai-Xizang Highway Based on Improved Remote Sensing Ecological Index].

The ecological environment along the Qinghai-Xizang highway is an important part of the construction of the ecological civilization in the Xizang region, and current research generally suffers from difficulties in data acquisition, low timeliness, and failure to consider the unique "alpine saline" environmental conditions in the study area due to the unique geographical environment of the Qinghai-Xizang plateau. Based on the GEE platform and the unique geographical environment of the study area, the remote sensing ecological index (RSEI) was improved, and a new saline remote sensing ecological index (SRSEI) applicable to the alpine saline region was constructed by using principal component analysis as an ecological environment quality evaluation index. The spatial distribution pattern and temporal variation trend of ecological environment quality along the Qinghai-Xizang Highway Nagqu-Amdo section were analyzed at multiple spatial and temporal scales using the ArcGIS 10.3 platform and geographic probes, and the driving mechanisms of eight control factors, including natural and human-made, on the spatial and temporal changes in SRSEI were investigated. The results showed that:① compared with RSEI, SRSEI was more sensitive to vegetation and had a stronger discriminatory ability in areas with sparse vegetation and severe salinization, which is suitable for ecological quality evaluation in alpine saline areas. ② The spatial scale of ecological environment quality in the study area had obvious geographical differentiation, and the areas with poor ecological quality were mainly concentrated in the northern Amdo County, whereas the areas with excellent and good quality grades were mainly distributed in the central-western and southeastern Nagqu areas. On the temporal scale, the ecological environment of the study area as a whole showed an improvement trend over 32 years, and the vegetation cover in the central-western and southeastern areas increased significantly, which had a strong improvement effect on the ecological environment. The improvement area was 1 425.98 km2, accounting for 99.82%. The mean value of SRSEI was 0.49, with an overall fluctuating upward trend and an average increase of 0.015 7 a-1. ③ The land use pattern was the most driving influence factor in the change of ecological environment quality in the study area, with an average q value of 0.157 6 over multiple years, and the influence of environmental factors was low. The multi-factor interaction results showed that the ecological environment in the study area was the result of multiple factors acting together, all factors had synergistic enhancement under the interaction, the influence of human factors was gradually increasing, and the interaction of the net primary productivity (NPP) of vegetation and land use pattern was the main interactive control factor of ecological environment quality in the study area. This study can provide a theoretical basis for ecological environmental protection and sustainable development along the Nagqu to Amdo section.

Predicting health outcomes with intensive longitudinal data collected by mobile health devices: a functional principal component regression approach.

BACKGROUND: Intensive longitudinal data (ILD) collected in near real time by mobile health devices provide a new opportunity for monitoring chronic diseases, early disease risk prediction, and disease prevention in health research. Functional data analysis, specifically functional principal component analysis, has great potential to abstract trends in ILD but has not been used extensively in mobile health research. OBJECTIVE: To introduce functional principal component analysis (fPCA) and demonstrate its potential applicability in estimating trends in ILD collected by mobile heath devices, assessing longitudinal association between ILD and health outcomes, and predicting health outcomes. METHODS: fPCA and scalar-to-function regression models were reviewed. A case study was used to illustrate the process of abstracting trends in intensively self-measured blood glucose using functional principal component analysis and then predicting future HbA1c values in patients with type 2 diabetes using a scalar-to-function regression model. RESULTS: Based on the scalar-to-function regression model results, there was a slightly increasing trend between daily blood glucose measures and HbA1c. 61% of variation in HbA1c could be predicted by the three preceding months' blood glucose values measured before breakfast (P < 0.0001, [Formula: see text]). CONCLUSIONS: Functional data analysis, specifically fPCA, offers a unique tool to capture patterns in ILD collected by mobile health devices. It is particularly useful in assessing longitudinal dynamic association between repeated measures and outcomes, and can be easily integrated in prediction models to improve prediction precision.

Combined use of principal component analysis/multiple linear regression analysis and artificial neural network to assess the impact of meteorological parameters on fluctuation of selected PM2.5-bound elements.

Based on the data of the State of Global Air (2020), air quality deterioration in Thailand has caused ~32,000 premature deaths, while the World Health Organization evaluated that air pollutants can decrease the life expectancy in the country by two years. PM2.5 was collected at three air quality observatory sites in Chiang-Mai, Bangkok, and Phuket, Thailand, from July 2020 to June 2021. The concentrations of 25 elements (Na, Mg, Al, Si, S, Cl, K, Ca, Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Ga, As, Se, Br, Sr, Ba, and Pb) were quantitatively characterised using energy-dispersive X-ray fluorescence spectrometry. Potential adverse health impacts of some element exposures from inhaling PM2.5 were estimated by employing the hazard quotient and excess lifetime cancer risk. Higher cancer risks were detected in PM2.5 samples collected at the sampling site in Bangkok, indicating that vehicle exhaust adversely impacts human health. Principal component analysis suggests that traffic emissions, crustal inputs coupled with maritime aerosols, and construction dust were the three main potential sources of PM2.5. Artificial neural networks underlined agricultural waste burning and relative humidity as two major factors controlling the air quality of Thailand.

Multivariate Analysis of Light-Activated SMOX Gas Sensors.

Chemoresistive gas sensors made from SnO2, ZnO, WO3, and In2O3 have been prepared by flame spray pyrolysis. The sensors' response to CO and NO2 in darkness and under illumination at different wavelengths, using commercially available LEDs, was investigated. Operation at room temperature turned out to be impractical due to the condensation of water inside the porous sensing layers and the irreversible changes it caused. Accordingly, for sensors operated at 70 °C, a characterization procedure was developed and proven to deliver consistent data. The resulting data set was so complex that usual univariate data analysis was intricate and, consequently, was investigated by correlation and principal component analysis. The results show that light of different wavelengths affects not only the resistance of each material, both under exposure to the target gases in humidity and in its absence, but also the sensor response to humidity and the target gases. It was found that each of the materials behaves differently under light exposure, and it was possible to identify conditions that need further investigations.

Raman spectroscopy combined with multivariate statistical algorithms for the simultaneous screening of cervical and breast cancers.

Breast and cervical cancers are becoming the leading causes of death among women worldwide, but current diagnostic methods have many drawbacks, such as being time-consuming and high cost. Raman spectroscopy, as a rapid, reliable, and non-destructive spectroscopic detection technique, has achieved many breakthrough results in the screening and prognosis of various cancer tumors. Therefore, in this study, Raman spectroscopy technology was used to diagnose breast cancer and cervical cancer. A total of 225 spectra were recorded from 87 patients with cervical cancer, 60 patients with breast cancer, and 78 healthy individuals. The obvious difference in Raman spectrum between the three groups was mainly shown at 809 cm-1 (tyrosine), 958 cm-1 (carotenoid), 1004 cm-1 (phenylalanine), 1154 cm-1 (ß-carotene), 1267 cm-1 (Amide III), 1445 cm-1 (phospholipids), 1515 cm-1 (ß-carotene), and 1585 cm-1 (C = C olefinic stretch). We used one-way analysis of variance for these peaks and demonstrated that they were significantly different. Then, we combined the detected Raman spectra with multivariate statistical calculations using the principal component analysis-linear discrimination algorithm (PCA-LDA) to discriminate between the three groups of collected serum samples. The diagnostic results showed that the model's accuracy, precision, recall, and F1 score of the model were 92.90%, 92.62%, 92.10%, and 92.36%, respectively. These results suggest that Raman spectroscopy can achieve ultra-sensitive detection of serum, and the developed diagnostic models have great potential for the prognosis and simultaneous screening of cervical and breast cancers.

HyperVein: A Hyperspectral Image Dataset for Human Vein Detection.

HyperSpectral Imaging (HSI) plays a pivotal role in various fields, including medical diagnostics, where precise human vein detection is crucial. HyperSpectral (HS) image data are very large and can cause computational complexities. Dimensionality reduction techniques are often employed to streamline HS image data processing. This paper presents a HS image dataset encompassing left- and right-hand images captured from 100 subjects with varying skin tones. The dataset was annotated using anatomical data to represent vein and non-vein areas within the images. This dataset is utilised to explore the effectiveness of dimensionality reduction techniques, namely: Principal Component Analysis (PCA), Folded PCA (FPCA), and Ward's Linkage Strategy using Mutual Information (WaLuMI) for vein detection. To generate experimental results, the HS image dataset was divided into train and test datasets. Optimum performing parameters for each of the dimensionality reduction techniques in conjunction with the Support Vector Machine (SVM) binary classification were determined using the Training dataset. The performance of the three dimensionality reduction-based vein detection methods was then assessed and compared using the test image dataset. Results show that the FPCA-based method outperforms the other two methods in terms of accuracy. For visualization purposes, the classification prediction image for each technique is post-processed using morphological operators, and results show the significant potential of HS imaging in vein detection.

A nucleic acid-based surface-enhanced Raman scattering of gold nanorods in N-gene integrated principal component analysis for COVID-19 detection.

A rapid, simple, sensitive, and selective point-of-care diagnosis tool kit is vital for detecting the coronavirus disease (COVID-19) based on the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) strain. Currently, the reverse transcriptase-polymerase chain reaction (RT-PCR) is the best technique to detect the disease. Although a good sensitivity has been observed in RT-PCR, the isolation and screening process for high sample volume is limited due to the time-consuming and laborious work. This study introduced a nucleic acid-based surface-enhanced Raman scattering (SERS) sensor to detect the nucleocapsid gene (N-gene) of SARS-CoV-2. The Raman scattering signal was amplified using gold nanoparticles (AuNPs) possessing a rod-like morphology to improve the SERS effect, which was approximately 12-15 nm in diameter and 40-50 nm in length. These nanoparticles were functionalised with the single-stranded deoxyribonucleic acid (ssDNA) complemented with the N-gene. Furthermore, the study demonstrates method selectivity by strategically testing the same virus genome at different locations. This focused approach showcases the method's capability to discern specific genetic variations, ensuring accuracy in viral detection. A multivariate statistical analysis technique was then applied to analyse the raw SERS spectra data using the principal component analysis (PCA). An acceptable variance amount was demonstrated by the overall variance (82.4 %) for PC1 and PC2, which exceeded the desired value of 80 %. These results successfully revealed the hidden information in the raw SERS spectra data. The outcome suggested a more significant thymine base detection than other nitrogenous bases at wavenumbers 613, 779, 1219, 1345, and 1382 cm-1. Adenine was also less observed at 734 cm-1, and ssDNA-RNA hybridisations were presented in the ketone with amino base SERS bands in 1746, 1815, 1871, and 1971 cm-1 of the fingerprint. Overall, the N-gene could be detected as low as 0.1 nM within 10 mins of incubation time. This approach could be developed as an alternative point-of-care diagnosis tool kit to detect and monitor the COVID-19 disease.

Deep reinforcement learning and convolutional autoencoders for anomaly detection of congenital inner ear malformations in clinical CT images.

Detection of abnormalities within the inner ear is a challenging task even for experienced clinicians. In this study, we propose an automated method for automatic abnormality detection to provide support for the diagnosis and clinical management of various otological disorders. We propose a framework for inner ear abnormality detection based on deep reinforcement learning for landmark detection which is trained uniquely in normative data. In our approach, we derive two abnormality measurements: Dimage and Uimage. The first measurement, Dimage, is based on the variability of the predicted configuration of a well-defined set of landmarks in a subspace formed by the point distribution model of the location of those landmarks in normative data. We create this subspace using Procrustes shape alignment and Principal Component Analysis projection. The second measurement, Uimage, represents the degree of hesitation of the agents when approaching the final location of the landmarks and is based on the distribution of the predicted Q-values of the model for the last ten states. Finally, we unify these measurements in a combined anomaly measurement called Cimage. We compare our method's performance with a 3D convolutional autoencoder technique for abnormality detection using the patch-based mean squared error between the original and the generated image as a basis for classifying abnormal versus normal anatomies. We compare both approaches and show that our method, based on deep reinforcement learning, shows better detection performance for abnormal anatomies on both an artificial and a real clinical CT dataset of various inner ear malformations with an increase of 11.2% of the area under the ROC curve. Our method also shows more robustness against the heterogeneous quality of the images in our dataset.