Fast Characterization Research Articles

Free‐listing and semantic knowledge: A diagnostic tool for detecting Alzheimer’s disease?

AbstractBackgroundA semantic memory impairment figures prominently among the manifestations of Alzheimer’s disease (AD). That impairment manifests itself relatively early in the disease trajectory with longitudinal studies showing evidence in patients with mild cognitive impairment and preclinical AD. We investigate the impact of AD on specific semantic categories integral to daily function: living kinds and artefacts.MethodWe relied on a “free‐listing task” as developed in cultural domain analysis, and assessed its relationship with other cognitive and functional tests in clinical use. We chose the written version for the elicitation of the free‐lists that we analyze here. A written format leads to the generation of longer lists than an oral elicitation. The participant pool was comprised of 19 AD patients and a control group, 15 cognitively normal age‐matched adults.ResultThe median number of items generated in each list for the patients was 12 for living things and 18 for man‐made objects, for the control group, 39 and 45, respectively. Controlling for age, education, estimate of premorbid intellectual ability, and sex, group membership (AD or control) predicts performance in the free‐listing in the two categories. Holding demographic variables constant, functional status is inversely related to the performance on free‐listing.ConclusionEven when provided the modality (written) to generate longer lists, individuals with AD still performed worse than controls. Our findings confirm previous observations identifying an impairment of semantic knowledge in AD. Free‐listing is effective at detecting this impairment. This easy‐to‐administer, cost‐effective tool could facilitate the fast preliminary characterization of AD in distinct cultural environments.

Droplet Digital PCR for Fast and Accurate Characterization of NF1 Locus Deletions: Confirmation of the Predominant Maternal Origin of Type-1 Deletions

Neurofibromatosis type-1 is a genetic disorder caused by loss-of-function variants in the tumor-suppressor NF1. Approximately 4% to 11% of neurofibromatosis type-1 patients have a NF1 locus complete deletion resulting from nonallelic homologous recombination between low copy repeats. Codeleted genes probably account for the more severe phenotype observed in NF1-deleted patients. This genotype-phenotype correlation highlights the need for a detailed molecular description. A droplet digital PCR (ddPCR) setalong the NF1 locus was designed to delimitate the three recurrent NF1 deletion breakpoints. The ddPCR was tested in 121 samples from nonrelated NF1-deleted patients. Classification based on ddPCR versus multiplex ligation-dependent probe amplification (MLPA) was compared. In addition, microsatellites were analyzed to identify parental origin of deletions. ddPCR identified 77 type-1 (64%), 20 type-2 (16%), 7 type-3 (6%), and 17 atypical deletions (14%). The results were comparable with MLPA, except for three atypical deletions misclassified as type-2 using MLPA, for which the SUZ12 gene was not deleted. A significant maternal bias (25 of 30) in the origin of deletions was identified. This study proposes a fast and efficient ddPCR quantification to allow fine NF1 deletion classification. It indicates that ddPCR can be implemented easily into routine diagnosis to complement the techniques dedicated to NF1 point variant identification. This new tool may help unravel the genetic basis conditioning phenotypic variability in NF1-deleted patients and offer tailored genetic counseling.

On-chip, non-invasive detection of all atrial arrhythmias for immediate and accurate characterization in adult rats

Abstract Background To monitor and record the occurrence and duration of cardiac arrhythmias in freely moving animals, ECG telemetry devices are often used. However, traditional telemetry devices require complex equipment unsuitable for on-site application in most animal housing facilities. Additionally, current atrial arrhythmias detection algorithms are mainly based on ECG R-R interval variability for atrial fibrillation (AF) detection. Such algorithms do not allow the detection of stable R-R interval atrial arrhythmias such as flutter. Thus there is currently a lack of non-invasive methods allowing automated detection of all atrial arrhythmias subtypes, thereby precluding in-depth research into general and subtype-specific mechanisms and treatment. Purpose Here we aim to create a non-invasive system allowing fast and accurate on-site detection and characterization of all atrial arrhythmias in rats. Methods Our atrial arrhythmia detector system was made of an on-skin ECG sensor, a low-power microprocessor unit, a large data storage unit and a battery. All components were assembled within a lightweight rat jacket. The filtered ECG signal was sent to the microprocessor allowing algorithm-based, real-time detection of atrial arrhythmias. The device allowed storage of ECG traces as well as arrhythmia characteristics (including type of arrhythmia and its duration). Atrial arrhythmias were induced by transesophageal atrial burst pacing in Wistar rats. These arrhythmia recordings were used to develop an AI detection method. Next, real-time detection algorithms based on R-R variability and AI were both integrated on-chip and compared using the same ECG data. Results ECG traces measured with our new system were similar to classical ECG measurements. Stored ECG traces provided the raw data for the development of an AI-based detection algorithm. The AI training was done using ECG data recorded from five different rats and reached 99.5% self-testing accuracy to detect all sub-types of atrial arrhythmias based on pre-categorized data. Using uncategorized data (n=25 from 5 rats) different from the training dataset, the AI algorithm could detect >80% of all types of atrial arrhythmias within 3s after initiation, while the rest could be detected within 10s. Though R-R interval based algorithm showed a high detection rate of 88% of AF within 10s, this method did not allow the detection of flutter or AF presenting regular R-R intervals within 1 minute after initiation. Conclusion Our atrial arrhythmia detection and analysis system provides a novel method for non-invasive investigation into atrial arrhythmias in rats and could easily be scaled for other species. It allows on-site fast and accurate detection and characterization of atrial arrhythmia subtypes without external equipment, thereby opening new possibilities for mechanistic studies and therapeutic testing, including closed-loop applications aiming for arrhythmia termination.

Super‐Resolution Photothermal Imaging at the Microscale by Model‐Based Image Reconstruction

Super‐resolution strategies expand the applicability of imaging modalities toward previously inaccessible spatial scales. Herein, photoactivated far‐infrared thermography is pushed from the millimeter to the micrometer scale by a 2D super‐resolution imaging approach capable of tackling the resolution barriers imposed by both diffraction‐limited signal collection and lateral spatiotemporal heat diffusion. The proposed imaging strategy relies on a full‐wave forward model of far‐infrared thermography and on an image‐inversion approach, which reconstructs the surface distribution of the sample photothermal absorbers via gradient‐descent optimization as the one yielding the best match between predicted and experimental images. With minute‐long acquisitions on a commercial low‐cost far‐infrared camera, less‐than‐5 μm spatial resolution is demonstrated on both synthetic samples and human liver biopsies ex vivo, thereby exemplifying the applicability of super‐resolution photothermal imaging to the fast nondestructive characterization of biological specimens well below the tissue spatial scale.

Characterization of long-term aged bitumen with FTIR spectroscopy and multivariate analysis methods

One of the most common techniques used to investigate bitumen aging from a chemical perspective is Fourier-Transformation Infrared (FTIR) spectroscopy. However, the majority of standard spectral interpretation simply covers the observation of single bands associated with carbonyls, sulfoxides or aromatic structures. A possibility for a more comprehensive post-processing of FTIR data is to apply multivariate analysis (MVA) methods. Although, MVA methods have already been used for classifying differently aged bitumen in the past, the deployed methods only included standard laboratory aging and failed to consider factors like reactive oxygen species or light. The purpose of this study was the investigation of aged bitumen samples with FTIR spectroscopy and subsequent data processing with MVA. Therefore, nine bituminous binders were subjected to aging with the Pressure Aging Vessel (PAV), Viennese Binder Aging (VBA) and light, followed by a characterization with FTIR spectroscopy and various MVA methods. It was possible to investigate the general structure of the acquired data and differentiate between the aging methods solely based on FTIR spectra. Furthermore, important wavenumbers for the classification were identified, including spectral features around the wavenumbers 3371, 3027, 2943, 1550 and 1324 and 848 cm−1 for the original spectra. This study shows the potential of MVA to maximize the information content gained by experimental analysis and for the development of fast and easy characterization methods.

Fast uncertainty quantification of spent nuclear fuel with neural networks

The accurate calculation and uncertainty quantification of the characteristics of spent nuclear fuel (SNF) play a crucial role in ensuring the safety, efficiency, and sustainability of nuclear energy production, waste management, and nuclear safeguards. State of the art physics-based models, while reliable, are computationally intensive and time-consuming. This paper presents a surrogate modeling approach using neural networks (NN) to predict a number of SNF characteristics with reduced computational costs compared to physics-based models. An NN is trained using data generated from CASMO5 lattice calculations. The trained NN accurately predicts decay heat and nuclide concentrations of SNF, as a function of key input parameters, such as enrichment, burnup, cooling time between cycles, mean boron concentration and fuel temperature. The model is validated against physics-based decay heat simulations and measurements of different uranium oxide fuel assemblies from two different pressurized water reactors. In addition, the NN is used to perform sensitivity analysis and uncertainty quantification. The results are in very good alignment to CASMO5, while the computational costs (taking into account the costs of generating training samples) are reduced by a factor of 10 or more. Our findings demonstrate the feasibility of using NNs as surrogate models for fast characterization of SNF, providing a promising avenue for improving computational efficiency in assessing nuclear fuel behavior and associated risks.

Extraction method impact on the physicochemical characteristics of lipids from chia nutlets applicable to long-term storage studies.

Lipids are relevant during the seed aging process, for which it is pertinent to choose an extraction method that does not alter their nature. Thus, three methods were applied to extract lipids from chia seeds: one used as reference (Soxhlet) and two at room temperature using hexane/ethanol (COBio) and hexane/isopropanol (COHar). The fatty acid composition and the tocopherol content of the oils were analyzed. Also, their oxidative status through the peroxide index, conjugated dienes and trienes, and malondialdehyde were determined. Besides, biophysical techniques, such as DSC and FT-IR, were applied. The extraction yield was not affected by the extraction method, while the fatty acid composition presented slight differences. Despite the high content of PUFAs, the oxidation level was low in all cases, especially in COBio, associated with the high content of α-tocopherol. DSC and FT-IR outcomes coincided with those obtained by conventional studies, resulting in efficient and fast characterization tools.

Fast and reliable characterization of egg triacylglycerol profiles using atmospheric solids analysis probe tandem mass spectrometry (ASAP-MS/MS) combined with chemometric tools

In this study, the potential of atmospheric solid analysis probe tandem mass spectrometry (ASAP-MS/MS) was assessed as a green and rapid method combined with gas chromatography with flame ionization detection (GC-FID) and chemometrics to determine the triacylglycerol (TAG) profile of eggs from various housing systems, including indoor, cage-free, free-range, organic, and enriched systems. The most abundant ions were identified using the product ion and neutral loss scan. The fatty acid composition, mainly unsaturated, palmitic, and stearic acids of eggs was influenced by the housing systems. The optimal conditions of ASAP-MS for cone voltage, desolvation gas temperature, and desolvation gas flow were determined to be 40 V, 600 °C, and 1100 L h−1, respectively. The main TAGs (50:2, 50:1, 52:4, 52:3, 52:2, 52:1, 54:4, 54:3, 54:2, 54:1) were observed in the mass range m/z 830–890. Notable differences were observed for TAGs 54:1 and 52:4, which were found in significantly higher levels on WE and CE. The proposed methodology enables rapid and efficient TAG profiling of egg yolks, providing a solid foundation for future research in this field.

Quantitative analysis of biopolymers in lignocellulosic biomass feedstocks via laser-assisted micro-pyrolysis flowing atmospheric-pressure afterglow high-resolution ambient mass spectrometry

Herein, a diode laser-assisted micro-pyrolysis (LAMP) technique coupled with FAPA high resolution mass spectrometry (HRMS) is demonstrated for fast chemical characterization of lignocellulosic biomass feedstocks. The solid lignocellulosic biomass can be analyzed directly with minimal sample preparation. The mass spectra of the pyrolysis products are interpreted with the aid of data visualization tools such as Kendrick mass defect (KMD) plots and van Krevelen plots. Furthermore, quantitation of lignin/cellulose/hemicellulose, sugar contents of glucan/xylan/galactan/arabinan and lignin monomeric unit S/G is achieved with good accuracy and precision, through multivariate analysis methods, including partial least squares regression (PLSR) and support vector regression (SVR).

Non-invasive assessment of steatohepatitis indicates increased risk of coronary artery disease.

Fatty liver diseases (FLD), especially defined as metabolic dysfunction-associated FLD (MAFLD), is of growing importance for patients and health-care providers. Extrahepatic comorbidities, predominantly coronary artery disease (CAD), contribute to excess morbidity and mortality in FLD. Although the association of FLD and CAD is well known, underlying pathophysiological links are not fully understood. Non-invasive means of liver diagnostic enable a fast and thorough characterization of FLD. We therefore assessed the severity of FLD in a cohort of patients at risk of CAD. Patients scheduled for coronary angiography were characterized by anthropometry, serum-based indices of liver fibrosis (NFS, FIB4), abdominal ultrasound and vibration controlled transient elastography (VCTE) including controlled attenuation parameter (CAP) and the Fibroscan-AST (FAST) score. Patients were stratified according to indication of therapeutic coronary intervention. 120 patients were recruited, MAFLD was found in 41%, while advanced fibrosis or cirrhosis were present in only 5%. Coronary vascular intervention was indicated in 42% (n = 50). Severity of steatosis assessed by CAP and risk of fibrosis defined by elevated liver stiffness (VCTE>8 kPa) and fibrosis indices were associated with the need for coronary intervention. FAST score, a marker of fibrotic steatohepatitis, was elevated in the intervention group (0.22 vs. 0.12, p<0.001). Multivariate regression analysis revealed FAST score as strongest predictor of CAD (OR 2.3 95%, CI 1.40-2.96). MAFLD is a frequent comorbidity in patients at CAD risk, but advanced liver disease has a low prevalence in patients undergoing elective coronary angiography. Therefore, a routine VCTE-based screening for FLD cannot be recommended in cardiac patients. The association of indicators of steatohepatitis with advanced CAD points to inflammatory processes as a conjoint mechanism of both diseases.

Integrated Microfluidic Chip for Rapid Antimicrobial Susceptibility Testing Directly from Positive Blood Cultures.

Rapid and accurate antimicrobial prescriptions are critical for bloodstream infection (BSI) patients, as they can guide drug use and decrease mortality significantly. The traditional antimicrobial susceptibility testing (AST) for BSI is time-consuming and tedious, taking 2-3 days. Avoiding lengthy monoclonal cultures and shortening the drug sensitivity incubation time are keys to accelerating the AST. Here, we introduced a bacteria separation integrated AST (BSI-AST) chip, which could extract bacteria directly from positive blood cultures (PBCs) within 10 min and quickly give susceptibility information within 3 h. The integrated chip includes a bacteria separation chamber, multiple AST chambers, and connection channels. The separator gel was first preloaded into the bacteria separation chamber, enabling the swift separation of bacteria cells from PBCs through on-chip centrifugation. Then, the bacteria suspension was distributed in the AST chambers with preloaded antibiotics through a quick vacuum-assisted aliquoting strategy. Through centrifuge-assisted on-chip enrichment, detectable growth of the phenotype under different antibiotics could be easily observed in the taper tips of AST chambers within a few hours. As a proof of concept, direct AST from artificial PBCs with Escherichia coli against 18 antibiotics was performed on the BSI-AST chip, and the whole process from bacteria extraction to AST result output was less than 3.5 h. Moreover, the integrated chip was successfully applied to the diagnosis of clinical PBCs, showing 93.3% categorical agreement with clinical standard methods. The reliable and fast pathogen characterization of the integrated chip suggested its great potential application in clinical diagnosis.

A paradigm shift in the prevention and diagnosis of oral squamous cell carcinoma.

Oral squamous cell carcinoma (OSCC) is a widespread disease with only 50%-60% 5-year survival. Individuals with potentially malignant precursor lesions are at high risk. Survival could be increased by effective, affordable, and simple screening methods, along with a shift from incisional tissue biopsies to non-invasive brush biopsies for cytology diagnosis, which are easy to perform in primary care. Along with the explainable, fast, and objective artificial intelligence characterisation of cells through deep learning, an easy-to-use, rapid, and cost-effective methodology for finding high-risk lesions is achievable. The collection of cytology samples offers the further opportunity of explorative genomic analysis. Our prospective multicentre study of patients with leukoplakia yields a vast number of oral keratinocytes. In addition to cytopathological analysis, whole-slide imaging and the training of deep neural networks, samples are analysed according to a single-cell RNA sequencing protocol, enabling mapping of the entire keratinocyte transcriptome. Mapping the changes in the genetic profile, based on mRNA expression, facilitates the identification of biomarkers that predict cancer transformation. This position paper highlights non-invasive methods for identifying patients with oral mucosal lesions at risk of malignant transformation. Reliable non-invasive methods for screening at-risk individuals bring the early diagnosis of OSCC within reach. The use of biomarkers to decide on a targeted therapy is most likely to improve the outcome. With the large-scale collection of samples following patients over time, combined with genomic analysis and modern machine-learning-based approaches for finding patterns in data, this path holds great promise.

Fast Diffusion Characterization by Multiphoton Excited Fluorescence Recovery while Photobleaching.

Multiphoton-excited fluorescence recovery while photobleaching (FRWP) is demonstrated as a method for quantitative measurements of rapid molecular diffusion over microsecond to millisecond timescales. Diffusion measurements are crucial in assessing molecular mobility in cell biology, materials science, and pharmacology. Optical and fluorescence microscopy techniques enable non-invasive rapid analysis of molecular diffusion but can be challenging for systems with diffusion coefficients exceeding ∼100 μm2/s. As an example, fluorescence recovery after photobleaching (FRAP) operates on the implicit assumption of a comparatively fast photobleaching step prior to a relatively slow recovery and is not generally applicable for systems exhibiting substantial recovery during photobleaching. These challenges are exacerbated in multiphoton excitation by the lower excitation efficiency and competing effects from local heating. Herein, beam-scanning FRWP with patterned line-bleach illumination is introduced as a technique that addresses FRAP limitations and further extends its application range by measuring faster diffusion events. In FRWP, the recovery of fluorescence is continuously probed after each pass of a fast-scanning mirror, and the upper bound of measurable diffusion rates is, therefore, only limited by the mirror scanning frequency. A theoretical model describing transient fluctuations in fluorescence intensity arising as a result of combined contributions from photobleaching and localized photothermal effect is introduced along with a mathematical framework for quantifying fluorescence intensity temporal curves and recovering room-temperature diffusion coefficients. FRWP is then tested by characterization of normal diffusion of rhodamine-labeled bovine serum albumin, green fluorescence protein, and immunoglobulin G molecules in aqueous solutions of varying viscosity.

Reliable Initial Model Selection for Efficient Characterization of Channel Reservoirs in Ensemble Kalman Filter

Abstract Ensemble Kalman filter is typically utilized to characterize reservoirs with high uncertainty. However, it requires a large number of reservoir models for stable and reliable update of its members, resulting in high simulation time. In this study, we propose a sampling scheme using convolutional autoencoder and principal component analysis for fast and reliable channel reservoir characterization. The proposed method provides good initial models similar to the reference model and gives successful model update for reliable quantification of future performances of channel reservoirs. Despite using fewer than 50 reservoir models, we achieve similar or even superior results compared to using all 400 initial models in this study. We demonstrate that the proposed scheme with ensemble Kalman filter provides faithful assimilation results while saving computation time.

Efficient mapping of lignin depolymerization product pools and quantification of specific monomers through rapid GPC-HPLC-UV/VIS analysis

BackgroundGrowing research on lignin depolymerization to functionalized bio-aromatics has necessitated dedicated analysis techniques. However, immense variability in molecular weight and functional groups of the depolymerization products impedes fast analysis of a large number of samples while remaining in-depth enough for catalyst screening or reaction condition optimization. While GPC-HPLC-UV/VIS has been a promising technique, up until now, the information it provides is largely qualitative. By enabling quantification of key monomeric products and through further reduction of overall analysis time, this study aims to increase the potential of GPC-HPLC-UV/VIS for fast and in-depth characterization of lignin depolymerization product pools. ResultsAnalysis of selected samples, isolated from GPC-HPLC-UV/VIS analyses of lignin depolymerization product pools, with gas chromatography (GC) equipped with an Orbitrap high-resolution accurate mass spectrometer (Orbitrap-HR/AM-MS) is successful in identifying the main low monomeric products. Moreover, these identifications are further substantiated through GPC-HPLC-UV/VIS analysis of standards. Furthermore, straight forward quantification of these products directly within GPC-HPLC-UV/VIS is successfully developed with limits of detection ≤0.05 mmol/L, which is at least on par with more complex analysis techniques. Additionally, several different reversed phase columns are assessed to reduce 2nd dimension (2D) time and, hence, overall analysis time while maintaining the possibility for quantification. A reduction in overall analysis time of about 30% as compared to the state-of-the-art is achieved by using a YMC Triart BIO C4 column as 2D. SignificanceThrough the enhancements introduced in this study, GPC-HPLC-UV/VIS emerges as a unique technique for the analysis of lignin depolymerization product pools, which is capable of fast yet sufficiently in-depth analysis of a large volume of samples. This capability is indispensable for catalyst screening and fine-tuning reaction conditions.

Undoped junctionless EZ-FET: Model and measurements

The junctionless EZ-FET is a device with simplified architecture and processing. With only two lithography levels and using regular process steps, it enables fast electrical characterization of semiconductor films on insulators (SOI) with undoped source and drain. An electrical model reproducing the peculiar transfer characteristics of a junctionless EZ-FET is presented in this work. Based on this model, a simplified parameter extraction methodology is proposed and used to access the electrical properties of SOI structures (i.e. threshold voltage, mobility).

Characterizing heterogenous microstructures of fiber-reinforced composite materials using an advanced image processing-based approach through optical microscopic images

The microstructures of heterogeneous fiber-reinforced composite materials exhibit high variations, which may affect the localized material mechanical properties, create the material machinability issue, and further cause severe surface integrity issues during machining processes. Most existing approaches to studying the heterogenous microstructure of composite usually require mechanical testing and extensive post-material characterizations. Even worse, experimental testing can damage material permanently, and offline characterizations (e.g., SEMs) are expensive and time-consuming. Therefore, creating a noncontact, fast, accurate, and affordable material characterization tool can help analyze heterogenous microstructures of composite materials and boost their industrial applications. This study presents an image process-based analytic approach to characterizing microstructural features. This approach performs well in extracting microstructures from blurred and translucent optical microscopic images. The further generated features, including fiber volume fraction and other physical descriptors, capture the microstructural features and quantify the material's heterogeneity. An exemplary case study using a two-dimension spatial pattern and an experimental case study show that the presented approach allows rapid and accurate characterizations of fiber-reinforced composites with anomaly detections on the local microstructural heterogeneity.

Transient Self-Discharge after Formation in Lithium-Ion Cells: Impact of State-of-Charge and Anode Overhang

A fast determination of cell quality after formation is challenging due to transient effects in the self-discharge measurement. This work investigated the self-discharge of NMC622/graphite single-layer pouch cells with varying anode dimensions to differentiate between SEI growth and anode overhang equalization processes. The transient self-discharge was measured directly after formation via voltage decay and for 20 weeks of calendar storage at three states-of-charge (SOC), 10%, 30%, and 50%. The transient behavior persisted for the entire measurement duration, even at a low SOC. Still, the low SOC minimized the impact of SEI growth and anode overhang equalization compared to moderate SOCs. Evaluating the coulombic efficiency from cycle aging showed a distinct capacity loss for the first cycle after storage, indicating further SEI growth, which stabilized in subsequent cycles. The aged capacity after cycling showed no significant dependence on the calendar storage, which further promotes fast self-discharge characterization at low SOC.

Numerical analysis and experimental verification of optical scattering from microplastics.

Accurate and fast characterization of the micron-sized plastic particles in aqueous media requires an in-depth understanding of light interaction with these particles. Due to the complexity of Mie scattering theory, the features of the scattered light have rarely been related to the physical properties of these tiny objects. To address this problem, we reveal the relation of the wavelength-dependent optical scattering patterns with the size and refractive index of the particles by numerically studying the angular scattering features. We subsequently present a low-cost setup to measure the optical scattering of the particles. Theoretical investigation shows that the angular distribution of the scattered light by microplastics carries distinct signatures of the particle size and the refractive index. The results can be used to develop a portable, low-cost setup to detect microplastics in water.

Machine learning based dual flat-spherical indentation approach for rough metallic surfaces

Instrumented indentation test (IIT) is regarded as one of the non-destructive test methods for mechanical characterization and safety assessment of engineering components. However, indentation test is very sensitive to surface topography i.e surface roughness, requiring specimens with ideally flat and smooth surface for producing reliable and reproducible test results. Recently, a dual flat-spherical indentation (DFSI) technique was proposed to extract material properties directly from rough metallic surface via indentation tests. Integration of DFSI technique with machine learning (ML) approach can facilitate reliable and fast characterization of rough metallic surfaces. Here, ML models including artificial neural network (ANN) and physics-informed artificial neural network (PI-ANN) are established, and then trained by using a database of indentation parameters generated via DFSI simulations. Spherical indentation techniques available in the literature are used to calculate the physics-informed loss in the PI-ANN model. The performances of trained ANN and PI-ANN models are compared, and an optimal ML model with hyperparameters is suggested based on validation study; PI-ANN model with sigmoid activation function shows a better performance than the ANN model and baseline model. The model performances are evaluated by performing DFSI experiments with SM45C and SS304. The applicability of ML based-DFSI method is demonstrated for extracting the mechanical properties from the rough surface of general metals.