BDBeni B. Dangi

Beni B. Dangi

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Beni B. Dangi has a total of 45 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Concentration Determination of Aspirin Deposit Patterns Using Quantitative Image Analysis and Artificial Intelligence

    by Sanam Pudasaini, Danielle O. Eniola, Oliver Steinbock, Beni B. Dangi

    The rapid detection of substandard or falsified pharmaceuticals remains a major challenge, particularly in regions with limited analytical infrastructure. In this study, a low-cost imaging-based method combined with machine learning was investigated for quantitative estimation of aspirin (acetylsalicylic acid) concentration through analysis of dried sessile droplet patterns. Aspirin solutions at concentrations ranging from 6% to 100% (saturation) were deposited on glass slides and allowed to evaporate under ambient conditions. The resulting crystallization patterns were imaged using a high-resolution camera, and 46 geometric and textural features were extracted from each image. Principal component analysis (PCA) revealed concentration-dependent clustering of deposit morphologies, indicating that structural differences in the dried patterns correlate with aspirin concentration. Regression models including Random Forest and XGBoost were trained using the extracted image features, while deep learning models (ResNet-50 and EfficientNet-B3) were applied for direct image-based regression. Amongst the models, EfficientNet-B3 achieved the highest predictive accuracy with an R2 = 0.99, while feature-based models also demonstrated strong predictive performance with R2 = 0.98. This approach was further evaluated using a commercial aspirin tablet sample, and results were compared with concentrations obtained through more conventional UV–visible spectroscopy using the standard addition method. This work demonstrates that droplet deposition imaging combined with artificial intelligence provides a promising, rapid, and non-destructive strategy for pharmaceutical concentration analysis and quality assessment.

  • Machine Learning-Enabled Image Analysis of Complex Chemical Mixtures: Synthetic Urine Droplets as a Test System

    by Amrutha S V, Beni B. Dangi, Oliver Steinbock

    Dried droplet deposits encode compositional information through evaporation-driven self-organization, offering a potential low-cost sensing modality for chemical analyses. Here, we evaluate the feasibility of inferring solute concentrations and pH from photographic images of dried synthetic urine stains by using automated experimentation and machine learning. A robotic droplet imager (RODI) combined with Latin hypercube sampling was employed to efficiently explore compositional space and generate standardized data sets comprising 70,000 images across two experimental systems: a two-component urea-NaCl model and seven-component synthetic urine mixtures spanning physiologically relevant ranges for urea, NaCl, KCl, creatinine, glucose, a combined salt group, and pH. Each image was reduced to 47 descriptors, capturing spatial morphology and texture. Random Forest regression yielded useful predictive performance for NaCl in the two-component system and identified pH as the most robustly inferred parameter in the complex mixture. However, continuous prediction of individual solute concentrations in the seven-component system remained limited, possibly due to the sparsity of training points across the high-dimensional compositional space. These findings delineate both the current capabilities and information-theoretic constraints of image-based compositional inference for complex chemical mixtures.

  • Open Access

    A Portable, Low‐Cost Spectrophotometer for Absorbance and Selected‐Wavelength Fluorescence Measurements

    by Sanam Pudasaini, Prakash Dangal, Rajesh Singh Rathore, Beni B. Dangi, Vijay Singh Chhetri

    We present a portable, low‐cost, modular spectrophotometric platform designed for wavelength‐selective optical measurements using interchangeable visible LEDs, an AS7343 multichannel spectral sensor, and Arduino‐based data acquisition. The system was developed to provide a compact alternative for optical absorbance and fluorescence measurements while maintaining performance comparable to conventional laboratory instrumentation. The device was evaluated using aqueous solutions of copper (II) sulfate (CuSO 4 ) and potassium permanganate (KMnO 4 ), and results were benchmarked against a standard UV–Vis spectrophotometer. The system demonstrated concentration‐dependent absorbance responses consistent with the Beer–Lambert law, with strong calibration linearity ( R 2 = 0.99 for CuSO 4 and R 2 = 0.97 for KMnO 4 ). In addition, the platform was adapted for fluorescence measurements of Rhodamine B via a simple reconfiguration to a 90° excitation–emission geometry. The fluorescence response showed linear behavior over the 0–4‐μM range, yielding an R 2 value of 0.97. The results indicate that the proposed system provides reliable optical measurements for both absorbance and fluorescence modalities, offering a flexible and cost‐effective alternative for analytical spectrophotometry applications in resource‐limited laboratory settings.

  • Open Access

    Synthetic images from generative AI for compositional analysis of dried solution patterns

    by Amrutha S V, Beni B. Dangi, Oliver Steinbock

    Image-based identification of solutes from dried drop stains offers a low-cost, scalable alternative to traditional chemical analysis, enabled by artificial intelligence and annotated image libraries. While recent advances in robotic droplet imaging have facilitated the collection of such datasets, experimental acquisition remains a key bottleneck. Here, we explore whether synthetic image generation can supplement or replace real data in the context of salt classification. Using Stable Diffusion with low-rank adaptation, we generate 133 000 synthetic images from as few as N = 1–128 experimental images per salt across seven common inorganic solutes. Synthetic images become visually indistinguishable from real ones for N ≥ 6; however, quantitative analysis based on 47 image metrics reveals subtle differences that vanish around N = 128. When used to train random forest, XGBoost, and multilayer perceptron (MLP) models, synthetic data alone achieve classification accuracies of up to 90% in some low-data regimes. Synthetic augmentation offers a powerful strategy for expanding training sets and enabling image-based chemical classification in data-scarce scenarios, even if it seldom outperforms models trained on real data. The realism of synthetic images also highlights the growing need for tools that detect AI-generated scientific images to ensure data integrity.


  • Morphological Signatures of Salt Crystals under Controlled Humidity Using Advanced Image Analysis

    by Sanam Pudasaini, Amrutha S. V., Oliver Steinbock, Beni B. Dangi

    We designed, fabricated, and utilized a humidity control chamber to investigate how controlled relative humidity (RH) affects crystallization patterns of two different salts; sodium chloride (NaCl) and ammonium chloride (NH4Cl) on glass slides. We also performed high-resolution imaging and MATLAB-based analysis to extract morphological and textural features of dried salt deposits, which were then assessed using Principal Component Analysis (PCA) technique. Results showed that humidity significantly impacted drying times and crystal morphologies. NH4Cl formed dendritic structures that grew more complex with higher humidity, while NaCl produced cubic/hopper crystals, whose size and aggregation varied with humidity. PCA analysis showed humidity-specific patterns, with NH4Cl displaying greater sensitivity. Our results demonstrate that the controlled humidity systematically alters salt crystallization dynamics, and advanced image analysis can precisely quantify these morphological signatures. Deep learning neural network models predict the correct salts from their image morphologies with over 97% accuracy.


  • Water Hardness from Photos of Dried Drop Stains: Reporter Salts Enable Low-Concentration Detection

    by Amrutha S V, Bruno C. Batista, Beni B. Dangi, Oliver Steinbock

    Water hardness is critical in domestic and industrial systems, typically measured by titration, atomic absorption spectroscopy, or inductively coupled plasma mass spectrometry. Here, we present a novel, photo-based method for quantifying water hardness using deposit patterns formed during the evaporation of hard water mixed with highly concentrated solutions of the reporter salts NaHCO3 and NaCl. This indirect approach enables the detection of hardness levels as low as 25 ppm, an improvement by a factor of 1,400 over earlier stain-based studies. Our method leverages high-resolution imaging, automated sample preparation, and advanced image analysis, including machine learning, to detect subtle morphological changes in the deposit patterns. A robotic drop imager (RODI) generated over 16,000 images, from which 47 geometric and textural metrics were extracted to train high-dimensional linear regression models and multilayer perceptron (MLP) networks. Using synthetic water samples with hardness levels of up to 250 ppm, we achieved prediction accuracies exceeding 91% for sodium bicarbonate-based patterns. Extending this methodology, we validated the methodology on real-world tap water, demonstrating its potential as a cost-effective alternative to traditional techniques and a versatile tool for water quality monitoring.


  • Understanding the Salt Crystallizations from Droplets under Various Gravity and Pressure Environments: Display of the Marangoni Effect?

    by Rim Hadidi, Venee D Pinckney, Sharlee A Shaw, Oliver Steinbock, Beni B Dangi

    Understanding the drying mechanisms, crystallization, and creep dynamics in salt solutions relies on a thorough understanding of the evaporation kinetics of individual droplets. These processes are of significant interest due to their implications in fields such as astrochemistry, environmental science, and material science. However, under nonequilibrium conditions like reduced pressure (g), they remain poorly understood, creating a pressing need for studies addressing these gaps. Our study investigates the evaporation of droplets from saturated salt solutions under nonequilibrium conditions, with experiments conducted in low-pressure and microgravity environments using a vacuum chamber and an acoustic levitation setup. We focused on the in situ crystallization dynamics of sodium chloride, potassium chloride, and ammonium chloride droplets, hypothesizing that the poleward migration of salt within levitating droplets is driven by the Marangoni effect, which arises from concentration gradients.1 Additionally, we explored the droplet evaporation mechanism under low pressure, examining factors contributing to the reduced "ring effect" in these conditions. Our findings indicate that the surface tension of the substrate under low pressure plays a crucial role in crystallization mechanisms, influencing the size of the central ring. These insights not only enhance our understanding of salt crystallization dynamics but also address the demand for innovative approaches to studying these processes in extraterrestrial and extreme environments. Furthermore, they provide a refined physical analysis and suggest potential applications in astrophysics, space exploration, and sustainable source management.