Correct Classification Rate Research Articles

Hazard functioning modeling for dextrose fluids associated hemodilution; Cox-regressional analysis for patients’ Hematocrit Hemoglobin Ratio

Aims: This study looks at the cox-regressional relationships between hematocrit to hemoglobin ratio (HHR) and a group of outcomes of interest (cOI). These outcomes include death from any cause, acute myocardial infarction, stroke, and severe dilutional hyponatraemia incidence. Additionally other outcome of interest in this study will be prolongation of admission days, infection rate, acute kidney injury, and mandating the transition to the critical unit and mechanical ventilation intubation. The study also will investigate the performances of the tested ratio and will explore their sensitivity, specificity, and accuracy indexes. Methods: Between 2018 and 2022, a retrospective study was conducted at the King Hussein Medical Centre in Amman, Jordan, involving 926 patients. The study aimed to determine the potential risk of dilution of red blood cells in circulation caused by dextrose-containing fluids in patients with chronic kidney disease. The study aimed to determine the ideal threshold levels of the investigated ratio (HHR) to enhance clinical outcomes and safety margins, particularly in patients with severe renal impairment or chronic kidney disease in a hospital setting. The study included patients with fluctuating kidney function, renal replacement therapy, uncontrolled diabetes, and conditions necessitating specific fluid regimens. The study used binary logistic regression and cox-regression proportional hazard modelling to examine the correlation between the hematocrit to haemoglobin ratio and the composite outcomes of interest (cOI) in patients from Jordan. The findings suggest that higher cOI likelihood indicates a more advantageous condition, while lower values suggest a less advantageous condition. Results: A retrospective study at the KHMC in Amman, Jordan, involved 1306 patients from 2018 to 2022. The study focused on the hematocrit to haemoglobin ratio (HHR) and achieved correct classification rates of 70.8%. The best HHR was 2.555:1, with a 42.8% chance of occurring at this level. Cox-Regression Proportional Hazard Modelling revealed a censored percentage of 83.4% in HHR>2.555 stratum and 51.5% in HHR≤2.555 stratum. The study found a high percentage of HD patients processed in the analysis. Conclusion: Intravenous dextrose fluids associated hemodilutional state is linked to all-cause mortality in admitted medically and surgically patients and cardiovascular death. Low ratio increases the risk. Factors such as malnutrition and kidney function loss contribute to prognosis. Future studies should focus on mechanistic pathways, optimal targets, age, and sociodemographic factors.

Differentiation of Δ9-THC and CBD Using Silver-Ligand Ion Complexation and Electrospray Ionization Tandem Mass Spectrometry (ESI-MS/MS).

The 2018 Farm Bill defines marijuana as Cannabis sativa L. or any derivative thereof that contains greater than 0.3% Δ9-tetrahydrocannabinol (Δ9-THC) on a dry weight basis. The main cannabinoids present in Cannabis sativa L., Δ9-THC and cannabidiol (CBD), are structural isomers that cannot be differentiated using direct mass spectrometry with soft ionization techniques alone. Due to the classification of marijuana as a Schedule I controlled substance, the differentiation of Δ9-THC and CBD is crucial within the seized drug community. This study explores the use of Ag-ligand ion complexation and electrospray ionization tandem mass spectrometry (ESI-MS/MS) for the differentiation of Δ9-THC and CBD using six different Ag complexes. Differences between the binding affinities of Δ9-THC and CBD for [Ag(PPh3)(OTf)]2 lead to the formation of unique product ions at m/z 421/423, m/z 353/355, and m/z 231 for CBD, enabling the differentiation of CBD from Δ9-THC. When applied to the analysis of known Δ9-THC:CBD mixture ratios, the developed [Ag(PPh3)(OTf)]2 ion complexation method was able to differentiate Δ9-THC-rich and CBD-rich samples based on the average abundance of the product ions at m/z 421/423. The developed approach was then applied to methanolic extracts of 20 authentic cannabis samples with known Δ9-THC and CBD compositions, resulting in a 95% correct classification rate. Even though the developed Ag-ligand ion complexation method was only demonstrated for the qualitative differentiation of Δ9-THC-rich and CBD-rich cannabis, this study establishes a foundation for the use of Ag-ligand ion complexation that is essential for future quantitative approaches.

Modeling interactions between Heparan sulfate and proteins based on the Heparan sulfate microarray analysis.

Heparan sulfate (HS), a sulfated polysaccharide abundant in the extracellular matrix, plays pivotal roles in various physiological and pathological processes by interacting with proteins. Investigating the binding selectivity of HS oligosaccharides to target proteins is essential, but the exhaustive inclusion of all possible oligosaccharides in microarray experiments is impractical. To address this challenge, we present a hybrid pipeline that integrates microarray and in silico techniques to design oligosaccharides with desired protein affinity. Using fibroblast growth factor 2 (FGF2) as a model protein, we assembled an in-house dataset of HS oligosaccharides on microarrays and developed two structural representations: a standard representation with all atoms explicit and a simplified representation with disaccharide units as "quasi-atoms." Predictive Quantitative Structure-Activity Relationship (QSAR) models for FGF2 affinity were developed using the Random Forest (RF) algorithm. The resulting models, considering the applicability domain, demonstrated high predictivity, with a correct classification rate of 0.81-0.80 and improved positive predictive values (PPV) up to 0.95. Virtual screening of 40 new oligosaccharides using the simplified model identified 15 computational hits, 11 of which were experimentally validated for high FGF2 affinity. This hybrid approach marks a significant step toward the targeted design of oligosaccharides with desired protein interactions, providing a foundation for broader applications in glycobiology.

Estimation of population affinity using proximal femoral measurements based on computed tomographic images in the Japanese and western Australian populations

The present study analyzes morphological differences femora of contemporary Japanese and Western Australian individuals and investigates the feasibility of population affinity estimation based on computed tomographic (CT) data. The latter is deemed to be of practical importance because most anthropological methods rely on the assessment of aspects of skull morphology, which when damaged and/or unavailable, often hampers attempts to estimate population affinity. The study sample comprised CT scans of 297 (146 females; 151 males) Japanese and 330 (145 females; 185 males) Western Australian adult individuals. A total of 10 measurements were acquired in two-dimensional CT images of the left and right femora; two machine learning methods (random forest modeling [RFM]) and support vector machine [SVM]) were then applied for population affinity classification. The accuracy of the two-way (sex-specific and sex-mixed) model was between 71.38 and 82.07% and 76.09–86.09% for RFM and SVM, respectively. Sex-specific (female and male) models were slightly more accurate compared to the sex-mixed models; there were no considerable differences in the correct classification rates between the female- and male-specific models. All the classification accuracies were higher in the Western Australian population, except for the male model using SVM. The four-way sex and population affinity model had an overall classification accuracy of 74.96% and 79.11% for RFM and SVM, respectively. The Western Australian females had the lowest correct classification rate followed by the Japanese males. Our data indicate that femoral measurements may be particularly useful for classification of Japanese and Western Australian individuals.

In Silico Prediction of Skin Sensitization for Compounds via Flexible Evidence Combination Based on Machine Learning and Dempster-Shafer Theory.

Skin sensitization is increasingly becoming a significant concern in the development of drugs and cosmetics due to consumer safety and occupational health problems. In silico methods have emerged as alternatives to traditional in vivo animal testing due to ethical and economic considerations. In this study, machine learning methods were used to build quantitative structure-activity relationship (QSAR) models on five skin sensitization data sets (GPMT, LLNA, DPRA, KeratinoSens, and h-CLAT), achieving effective predictive accuracies (correct classification rates of 0.688-0.764 on test sets). To address the complex mechanisms of human skin sensitization, the Dempster-Shafer theory was applied to merge multiple QSAR models, resulting in an evidence-based integrated decision model. Various evidence combinations and combination rules were explored, with the self-defined Q3 rule showing superior balance. The combination of evidence such as GPMT and KeratinoSens and h-CLAT achieved a correct classification rate (CCR) of 0.880 and coverage of 0.893 while maintaining the competitiveness of other combinations. Additionally, the Shapley additive explanations (SHAP) method was used to interpret important features and substructures related to skin sensitization. A comparative analysis of an external human test set demonstrated the superior performance of the proposed method. Finally, to enhance accessibility, the workflow was implemented into a user-friendly software named HSkinSensDS.

Agile scores in MASLD and ALD: External validation and their utility in clinical algorithms

Agile scores, including liver stiffness measurements (LSM) and routine clinical/laboratory biomarkers, have been developed for advanced fibrosis (F≥3) and cirrhosis (F4), respectively, in patients with metabolic dysfunction-associated steatotic liver disease (MASLD). We independently validated the diagnostic accuracy of these scores in MASLD, alcohol-related liver disease (ALD) and chronic hepatitis B or C (CHB/C) and assessed them in clinical algorithms with FIB-4 and LSM. We included 4,243 patients (MASLD: 912, ALD: 386, CHB: 597, CHC: 2,348) with LSM, liver biopsy and laboratory tests within 6 months. FIB-4, Agile 3+ and Agile 4 scores were calculated. For F≥3, the diagnostic accuracy of Agile 3+ and LSM were similar in MASLD (AUC: 0.86 vs. 0.86, p= 0.831) and ALD (0.92 vs. 0.94, p= 0.123). For cirrhosis, Agile 4 was similar to LSM in MASLD (0.89 vs. 0.90, p= 0.412) and ALD (0.94 vs. 0.95, p= 0.513). Agile 3+/4 performed worse than LSM in CHB/C. Using predefined dual thresholds of 90% sensitivity/specificity, correct classification rates in MASLD were 66% vs. 61% using Agile 3+ vs. LS dual cut-offs and 71% vs. 67% in ALD, respectively. When using Agile 3+ or LSM as a second step after FIB-4 >1.3, correct classification rates were higher with Agile 3+ than LSM, both for MASLD (75% vs. 71%) and ALD (76% vs. 72%), with fewer indeterminate results. Positive agreement of LSM and Agile 3+/4 significantly increased the specificity of a diagnosis of advanced fibrosis/cirrhosis. Agile 3+ and Agile 4 have equal diagnostic accuracy with LSM in both MASLD and ALD but result in fewer indeterminate results. Sequential use of FIB-4 and Agile 3+/4 or concurrent Agile 3+/4 and LSM can be used to further optimize F≥3 diagnosis. As of today, it is accepted that there will be no single non-invasive test or an isolated cut-off for identifying patients with advanced chronic liver disease. Here, we confirmed that Agile 3+ and Agile 4 scores are useful alternatives to simple liver stiffness measurement in diagnosing advanced fibrosis/cirrhosis in steatotic liver disease, but they do not perform as well in chronic viral hepatitis. Agile scores can help optimize the diagnosis of advanced fibrosis/cirrhosis in a dual cut-off strategy by reducing the number of indeterminate results either alone or in a sequential strategy after FIB-4. The combination of Agile scores and liver stiffness measurement can further increase our confidence in a positive diagnosis of advanced fibrosis/cirrhosis. These novel combination strategies can be useful tools to predict the likelihood of advanced stages of liver disease with the highest possible accuracy in a secondary/tertiary healthcare setting.

تحليل مقارن لخوارزميات التعلم الآلي لتصنيف مرض السكري باستخدام تحليل مصفوفة الارتباك

يستخدم العاملين في الرعاية الصحية التعلم الآلي أكثر فأكثر في السنوات الأخيرة لتعزيز نتائج المرضى وخفض التكاليف عليهم. بالإضافة إلى ذلك، تم تنفيذ التعلم الآلي في مجالات مختلفة، بما في ذلك تشخيص الأمراض، وتصنيف مخاطر المرضى، واقتراحات العلاج المخصصة، وتطوير الأدوية. يمكن لخوارزميات التعلم الآلي أن تفحص كميات هائلة من البيانات من السجلات الصحية الإلكترونية، والصور الطبية، ومصادر أخرى لتحديد الأنماط والتنبؤات، والتي يمكن أن تدعم المتخصصين في الرعاية الصحية والخبراء في اتخاذ قرارات مستنيرة، وتعزيز رعاية المرضى، وتحديد حالة المريض الصحية. في هذا الصدد، اختار المؤلف مقارنة أداء ثلاث خوارزميات (الانحدار اللوجستي ، Adaboost ، بايزالساذجة) من خلال معدل التصنيف الصحيح للتنبؤ بمرض السكري من أجل ضمان فعالية التشخيص الدقيق. تم الحصول على مجموعة البيانات المطبقة في هذا العمل من مستودع مؤسسي بجامعة فاندربيلت وهي بيانات متاحة للحميع. استنجتت الدراسة أن ثلاث خوارزميات فعالة للغاية في التنبؤ. بشكل أساسي، كان معدل تصنيف الانحدار اللوجستي و Adaboost أعلى من 92٪ ، وحققت خوارزمية بايز الساذجة معدل تصنيف أعلى من 90٪.

A sparse data gas sensor array feature mining method for rubber Mooney viscosity measurement

Mooney viscosity is an important index to reflect the performance and quality of rubber. At present, the rubber Mooney test has the problems of large time delay, destructive and unable to detect online, which restricts the development of rubber industry. In this paper, a gas sensor array-based online inspection method for rubber Mooney viscosity is proposed to improve the problems of measurement delay and raw material waste in the traditional method. A multiple generator time series generative adversarial network (MGTSGAN) structure is proposed to address the problem that the lack of sample data volume and uneven data distribution make it difficult to model. Transformer is introduced to solve the problem of traditional generative adversarial networks in dealing with long sequential dependencies. In the experimental part, a rubber Mooney viscosity detection device is built to verify the effectiveness of the proposed method. The performance of different generative models on two gas sensor datasets is compared to verify the advancement and generalization of the proposed method. The experimental results show that the correct rate of this paper’s method for rubber Mooney viscosity classification is higher than 96%. The correct rates are all improved after data enhancement, in which the MGTSGAN proposed in this paper obtains the highest correct rate of 98.35%. For the classification experiments on Gas sensor array under flow modulation dataset also achieved relatively good results. Among them, the highest correct classification rate of 95.63% is achieved after data enhancement using MGTSGAN.

Spectral Profiling (Fourier Transform Infrared Spectroscopy) and Machine Learning for the Recognition of Milk from Different Bovine Breeds.

The Podolica cattle breed is widespread in southern Italy, and its productivity is characterized by low yields and an extraordinary quality of milk and meats. Most of the milk produced is transformed into "Caciocavallo Podolico" cheese, which is made with 100% Podolica milk. Fourier Transform Infrared Spectroscopy (FTIR) is the technique that, in this research work, was applied together with machine learning to discriminate 100% Podolica milk from contamination of other Calabrian cattle breeds. The analysis on the test set produced a misclassification percentage of 6.7%. Among the 15 non-Podolica samples in the test set, 2 were misclassified and recognized as Podolica milk even though the milk was from other species. The correct classification rate improved to 100% when the same method was applied to the recognition of Podolica and Pezzata Rossa milk produced by the same farm. Furthermore, this technique was tested for the recognition of Podolica milk mixed with milk from other bovine species. The multivariate model and the respective confusion matrices obtained showed that all the 14 Podolica samples (test set) mixed with 40% non-Podolica milk were correctly classified. In addition, Pezzata Rossa milk produced by the same farm was detected as a contaminant in Podolica milk from the same farm down to concentrations as little as 5% with a 100% correct classification rate in the test set. The method described yielded higher accuracy values when applied to the discrimination of milks from different breeds belonging to the same farm. One of the reasons for this phenomenon could be linked to the elimination of the environmental variable. However, the results obtained in this work demonstrate the possibility of using FTIR to discriminate between milks from different breeds.

Schneiderian First Rank Symptoms Significantly Predict a Dissociative Disorder Diagnosis in Psychiatric In-Patients

ABSTRACT Previous empirical studies on the relationship between psychotic symptoms and dissociative disorders focused on auditory hallucinations only or employed limited statistical analyses. We investigated whether the frequency of Schneiderian first rank symptoms (FRS) predicts the presence or absence of a dissociative disorder (DD). Psychiatric in-patients (n = 116) completed measures of dissociation, FRS and general psychological distress (GPD). DD diagnoses were confirmed by multidisciplinary teams or administering the Structured Clinical Interview for DSM-IV Dissociative Disorders-Revised (SCID-D-R). The FRS were recorded in the Multidimensional Inventory of Dissociation (MID) and a mean score obtained for 35 relevant items: Voices arguing, voices commenting, made feelings, made impulses, made actions, influences on body, thought withdrawal, and thought insertion. A global severity index (GSI) of GPD was obtained from the Symptom Checklist–90–Revised (SCL-90-R). Logistic regression models examined whether FRS predict diagnostic classification of patients under a DD (n = 16) or not (n = 100), controlling for GSI. The overall fit of the model was significant (p = .0002). DD was correctly classified using frequency of FRS, controlling for GSI. The latter was moderately associated with FRS (r = 0.56). FRS more than doubled the odds of a DD diagnosis (odds = 2.089; 95% CI = 1.409–3.098; correct classification rate 87.1%). The study provides convincing evidence that FRS are closely related to DDs. FRS should alert clinicians to consider DDs in differential diagnosis of psychiatric in-patients. Future research should analyze whether FRS also predict a diagnosis of schizophrenia or other psychiatric disorders.

Enhancing Air Quality Monitoring Through Spatial Classification in Central Zone of Peninsular Malaysia Using Artifical Neural Network as Prediction Tools

Ambient air monitoring plays a crucial role in the effective implementation of air quality management systems. This practice entails systematically and over a long period assessing and quantifying specific pollutants in the outdoor environment. However, the high cost of acquiring sufficient equipment for comprehensive air monitoring poses a challenge. Thus, this study proposes that spatial classification could be a viable approach to reducing monitoring stations while still obtaining adequate air monitoring data. The objective of this study was to examine the predictive performance of artificial neural networks (ANNs) in spatial classification to support air quality monitoring. By implementing ANN in this study, the MLP-FF-ANN model successfully distinguished air quality samples in the HPC, MPC, and LPC regions. Particularly notable were the positive outcomes achieved with a configuration of ten hidden nodes, resulting in an R2 value of 0.7982, as well as the lowest RMSE (0.3799) and MR (0.1950). Additionally, the MLP-FF-ANN model demonstrated commendable performance, achieving an average correct classification rate of 76.38%. These findings suggest that air quality monitoring based on clustered data can effectively maintain data quality.

Genres Classification of Popular Songs Listening by Using Keras

Listening to the music affects the brain in ways which might help to promote the human health and arrange various diseases symptoms. Music is a phenomenon that is intertwined at every stage of human life. In the modern era music is shaped by the combination of an incredible number of genres, some of which are contemporary, and some come from the previous times. The music genre represents a collection of musical works that develop according to a certain shape, expression and technique. The music genre of interest varies from person to person in society. Most listeners today do not know what kind of music they listen to. In this study, sound features were extracted from music data and the Keras model was trained using these attributes. The correct classification rate of a music genre of the trained model was determined as 71.66%. Mel Frequency Cepstral Coefficients (MFCC), Mel Spectrogram, Chroma Vector and Tonnetz methods in the Librosa library were used to extract sound properties from music data. Using the features probed by means of the library, the most listened songs with Shazam in Türkiye were categorized in with TensorFlow/Keras. Many methods can be used in classification. It is uncertain which method the researchers should opt. It has been emphasized that classification of the genres of newly released songs by using Keras in this study. At result, it is said that the study has presented a sound processing are Keras classification of musical parts.

Predicting Hydrocarbon Primary Biodegradation in Soil and Sediment Systems Using System Parameterization and Machine Learning.

Technical complexity associated with biodegradation testing, particularly for substances of unknown or variable composition, complex reaction products, or biological materials (UVCB), necessitates the advancement of non-testing methods such as quantitative structure-property relationships (QSPRs). Models for describing the biodegradation of petroleum hydrocarbons (HCs) have been previously developed. A critical limitation of available models is their inability to capture the variability in biodegradation rates associated with variable test systems and environmental conditions. Recently, the Hydrocarbon Biodegradation System Integrated Model (HC-BioSIM) was developed to characterize the biodegradation of HCs in aquatic systems with the inclusion of key test system variables. The present study further expands the HC-BioSIM methodology to soil and sediment systems using a database of 2195 half-life (i.e., degradation time [DT]50) entries for HCs in soil and sediment. Relevance and reliability criteria were defined based on similarity to standard testing guidelines for biodegradation testing and applied to all entries in the database. The HC-BioSIM soil and sediment models significantly outperformed the existing biodegradation HC half-life (BioHCWin) and virtual evaluation of chemical properties and toxicities (VEGA) quantitative Mario Negri Institute for Pharmacological Research (IRFMN) models in soil and sediment. Average errors in predicted DT50s were reduced by up to 6.3- and 8.7-fold for soil and sediment, respectively. No significant bias as a function of HC class, carbon number, or test system parameters was observed. Model diagnostics demonstrated low variability in performance and high consistency of parameter usage/importance and rule structure, supporting the generalizability and stability of the models for application to external data sets. The HC-BioSIM provides improved accuracy of Persistence categorization, with correct classification rates of 83.9%, and 90.6% for soil and sediment, respectively, demonstrating a significant improvement over the existing BioHCWin (70.7% and 58.6%) and VEGA (59.5% and 18.5%) models. Environ Toxicol Chem 2024;43:1352-1363. © 2024 Concawe. Environmental Toxicology and Chemistry published by Wiley Periodicals LLC on behalf of SETAC.

Trace the origin of yak meat in Xizang based on stable isotope combined with multivariate statistics

In this study, the feasibility of tracing the origin of yak meat in Xizang Autonomous Region based on stable isotope combined with multivariable statistics was researched. The δ13C, δ15N, δ2H and δ18O in yak meat were determined by stable isotope ratio mass spectrometry, and the data were analyzed by analysis of variance, fisher discriminant analysis (FDA), back propagation (BP) neural network and orthogonal partial least squares discrimination analysis (OPLS-DA). The results showed that the δ13C, δ15N, δ2H and δ18O had significant differences among different origins (P < 0.05). The overall original correct discrimination rate of fisher discriminant analysis was 89.7 %, and the correct discrimination rate of cross validation was 88.2 %. The correct classification rate of BP neural network based on training set was 93.38 %, and the correct classification rate of BP neural network based on test set was 89.83 %. The OPLS-DA model interpretation rate parameter R2Y was 0.67, the model prediction rate parameter Q2 was 0.409, which could distinguish yak meat from seven different producing areas in Xizang Autonomous Region. The results showed that the origin of yak meat in Xizang Autonomous Region can be traced based on stable isotope combined with multivariate statistics.

Fluorescence spectral analysis and logistic regression modeling for diagnosing basal cell carcinoma on head and neck

The optical fluorescence method is distinguished by key features such as non-invasiveness, high sensitivity, and resolution, which are superior to traditional diagnostic approaches. Unlike histopathological examinations and biochemical analyses, optical diagnostic methods obviate the need for tissue sampling, enabling the analysis of virtually unlimited material.The research aims to examine the effectiveness of emission spectra analysis in the diagnosis of basal cell carcinoma (BCC) of the scalp and neck. The analysis was based on data provided by Specialized Hospital No. 2 in Bytom comprising a study sample of 10 patients. For each patient, fluorescence emission spectra were recorded from each of 512 points along a 5 mm line.The results obtained from the histopathological examination, the analysis and morphological evaluation of the tissue, and the diagnosis through microscopic observation were used to define a dichotomous variable (presence or absence of a cancerous lesion), adopted in the study as the modeled variable. The next step of the presented study involved constructing a logistic regression model for identifying cancerous lesions depending on the biochemical indicator's relative fluorescence value (RFV) and emission wavelength (ELW) within the 620 nm to 730 nm range. This wavelength range is often used in fluorescence diagnostics to detect various pathologies, including cancerous lesions.The resulting binary logistic regression model, logit(p)=−33.17+0.04ELW+0.01RFV, indicates a statistically significant relationship between wavelength and relative fluorescence values with the probability of detecting cancer. The estimated model exhibits a good fit and high predictive accuracy. The overall model accuracy is 84.8 %, with the correct classification rates at approximately 96 % for healthy individuals and 74 % for individuals with cancer. These findings underscore the potential of photodynamic diagnostics in cancer detection and monitoring.

Geographical discrimination of Paw San rice cultivated in different regions of Myanmar using near-infrared spectroscopy, headspace-gas chromatography-ion mobility spectrometry and chemometrics

Paw San rice, also known as “Myanmar pearl rice”, is considered the highest quality rice in Myanmar. There are considerable differences in terms of the premium commercial value of Paw San rice, which is an incentive for fraud, e.g. adulteration with cheaper rice varieties or mislabelling its geographical origin. Shwe Bo District is one of the most popular rice growing areas in the Sagaing region of Myanmar which produces the most valued and highly priced Paw San rice (Shwe Bo Paw San). The verification of the geographical origin of Paw San rice is not readily undertaken in the rice supply chain because the existing analytical approaches are time-consuming and expensive. Therefore, there is a need for rapid, robust and cost-effective analytical techniques for monitoring the authenticity and geographical origin of Paw San rice.In this 4-year study, two rapid screening techniques, Fourier-transform near-infrared (FT-NIR) spectroscopy and headspace-gas chromatography-ion mobility spectrometry (HS-GC-IMS), coupled with chemometric modelling, were applied and compared for the regional differentiation of Paw San rice. In addition, low-level fusion of the FT-NIR and HS-GC-IMS data was performed and its effect on the discriminative power of the chemometric models was assessed. Extensive model validation, including the validation using independent samples from a different production year, was performed. Furthermore, the effect of the sample preparation technique (grinding versus no sample preparation) on the performance of the discriminative model, obtained with FT-NIR spectral data, was assessed. The study discusses the suitability of FT-NIR spectroscopy, HS-GC-IMS and the combination of both approaches for rapid determination of the geographical origin of Paw San rice.The results demonstrated the excellent potential of the FT-NIR spectroscopy as well as HS-GC-IMS for the differentiation of Paw San rice cultivated in two distinct geographical regions. The OPLS-DA model, built using FT-NIR data of rice from 3 production years, achieved 96.67% total correct classification rate of an independent dataset from the 4th production year. The DD-SIMCA model, built using FT-NIR data of ground rice, also demonstrated the highest performance: 94% sensitivity and 97% specificity. This study has demonstrated that FT-NIR spectroscopy can be used as an accessible, rapid and cost-effective screening tool to discriminate between Paw San rice cultivated in the Shwe Bo and Ayeyarwady regions of Myanmar.

Tracing the geographic origin of CTC black tea based on colorimetric sensor array response to taste substances combined with chemometrics

The quality of crush–tear–curl black tea (CTC-BT) varies greatly by geographic origin. Origin traceability is crucial for consumer interest protection, market order regulation, and food safety monitoring. This paper proposes a fast and accurate method for qualitative discrimination of CTC-BT origins and quantitative detection of its key taste-presenting substances. The method involves a simple colorimetric sensor array and ultraviolet–visible spectroscopy. The effects of various variable screening methods on modeling results were compared. A particle swarm optimization–based support vector machine achieved the highest performance in qualitative discrimination, with a correct classification rate of 99.48%. Based on their origin-distinguishing contributions and dose-over-thresholds, seven key taste-presenting substances were screened, namely, theaflavin, caffine, vitexin-2-O-rhamnoside, rutin, epigallocatechin gallate, epicatechin gallate, gallic acid. A least squares-support vector regression model achieved accurate quantification of the seven aforementioned compounds (square root of determination coefficient of prediction >0.9698, residual prediction deviation >2).

Full Characterisation of Heroin Samples Using Infrared Spectroscopy and Multivariate Calibration.

The analysis of heroin samples, before use in the protected environment of user centra, could be a supplementary service in the context of harm reduction. Infrared spectroscopy hyphenated with multivariate calibration could be a valuable asset in this context, and therefore 125 heroin samples were collected directly from users and analysed with classical chromatographic techniques. Further, Mid-Infrared spectra were collected for all samples, to be used in Partial Least Squares (PLS) modelling, in order to obtain qualitative and quantitative models based on real live samples. The approach showed that it was possible to identify and quantify heroin in the samples based on the collected spectral data and PLS modelling. These models were able to identify heroin correctly for 96% of the samples of the external test set with precision, specificity and sensitivity values of 100.0, 75.0 and 95.5%, respectively. For regression, a root mean squared error of prediction (RMSEP) of 0.04 was obtained, pointing at good predictive properties. Furthermore, during mass spectrometric screening, 10 different adulterants and impurities were encountered. Using the spectral data to model the presence of each of these resulted in performant models for seven of them. All models showed promising correct-classification rates (between 92 and 96%) and good values for sensitivity, specificity and precision. For codeine and morphine, the models were not satisfactory, probably due to the low concentration of these impurities as a consequence of acetylation. For methacetin, the approach failed.

SUPER-COUGH: A Super Learner-based ensemble machine learning method for detecting disease on cough acoustic signals

Sound classification has obtained considerable attention in recent years due to its wide range of applications in various fields, such as speech recognition, sound surveillance, music analysis, and environmental monitoring. Because of its success, audio classification can also be employed in medical applications. Coughing is the most common disease symptom, and cough sounds might be used to diagnose them. This research focuses on identifying observable features of cough and classifying them into positive, negative, or symptomatic categories. A novel ensemble learning model based on the super learner (SL) is proposed to diagnose the disease using cough sounds utilizing various audio features such as Frequency Distribution, Time Domain Features, Spectral Features, and Time–Frequency Features. The SL method is a cross-validated approach to stacked generalization, and it can select an optimal learner from a set of learners and improve performance by selecting and merging models using cross-validation. The proposed SL model comprises DT, RF, LR, SVM, ET, and k-NN algorithms. We use the public Coughvid dataset, and the proposed model achieves a correct classification rate for symptomatic cases, which was 90.90%, and the positive predictive value for COVID-19 cases was 84.50%. The SL3 model attains 72%, 78%, 73%, 74.4%, and 78.85% precision, recall, f1-score, accuracy, and average AUC values, respectively. The numerical results show that the proposed model might be implemented to diagnose various other diseases that can be determined from respiratory sounds.

Predicting Failure Modes of RC Beams Strengthening with FRP-NSM System: A Statistical Approach

FRP strengthening using near surface mounted (NSM)is recognized as a highly effective method for strengthening structures using FRP material. However, there are some drawbacks associated with this technique. One of the challenges is concrete cover separation failure, which leads to premature material failure and restricts NSM technique applicability. To address this issue, multinomial logistic regression (MLR) analysis is used. This analysis aimed to investigate the factors (Total Equivalent Steel Ratio TESR%, Stirrup Steel Ratio%, and presence of an anchorage) that can be used to predict the failure modes of reinforced concrete beams strengthened with the FRP-NSM system using a previously published dataset consisting of 131 beam tests from 25 studies. By using odds ratios, which are exponential coefficients, the MLR results were interpreted. The study indicated that a reduction in the TESR% predictor or increase in Stirrup Steel Ratio%, deceases the probability of conversion from concrete cover separation to other failure modes such as flexural failure, FRP bar debonding, and FRP bar rupture. Moreover, the reinforced concrete beams strengthened with the FRP-NSM system with anchorage has a positive contribution on the expected the failure modes: flexural failure, FRP bar debonding, and FRP bar rupture compared to concrete cover separation. Based on the classification table analysis, the model’s correct classification rate was 51.1%. The model is accurate in predicting failure modes, as its rate is higher than the chance accuracy rate of 36.4%.