Dried biofluid droplet morphologies for automated and scalable disease classification

Anusuya Pal; Piyush Thakur; Amalesh Gope; Ryuta Uraki; Kawaoka Yoshihiro; Miho Yanagisawa

doi:10.1002/dro2.70054

Droplet ›› 2026, Vol. 5 ›› Issue (2) :e70054 DOI: 10.1002/dro2.70054

RESEARCH ARTICLE

Dried biofluid droplet morphologies for automated and scalable disease classification

Anusuya Pal ¹
, Piyush Thakur ²
, Amalesh Gope ³
, Ryuta Uraki ⁴^,⁵^,⁶
, Kawaoka Yoshihiro ⁴^,⁵^,⁶^,⁷
, Miho Yanagisawa ¹^,⁸^,⁹^,¹⁰

Author information +

¹. Graduate School of Arts and Sciences, The University of Tokyo, Komaba, Tokyo, Japan

². CSIR-Central Scientific Instruments Organisation, Chandigarh, India

³. Department of Linguistics and Language Technology, Tezpur University, Tezpur, Assam, India

⁴. The University of Tokyo Pandemic Preparedness, Infection and Advanced Research Center (The UTOPIA Center), Institutes for Advanced Study, The University of Tokyo, Bunkyo, Tokyo, Japan

⁵. The Research Center for Global Viral Diseases, National Center for Global Health and Medicine Research Institute, Shinjuku, Tokyo, Japan

⁶. Division of Virology, Institute of Medical Science, The University of Tokyo, Minato, Tokyo, Japan

⁷. Influenza Research Institute, Department of Pathobiological Sciences, School of Veterinary Medicine, University of Wisconsin-Madison, Madison, Wisconsin, USA

⁸. Komaba Institute for Science, Graduate School of Arts and Sciences, The University of Tokyo, Komaba, Tokyo, Japan

⁹. Center for Complex Systems Biology, Universal Biology Institute, The University of Tokyo, Komaba, Tokyo, Japan

¹⁰. Department of Physics, Graduate School of Science, The University of Tokyo, Hongo, Tokyo, Japan

apal@g.ecc.u-tokyo.ac.jp

Show less

History +

PDF (6557KB)

Abstract

Dried biofluid droplets contain complex physicochemical information, yet their diagnostic potential remains largely limited due to the absence of an accessible, end-to-end analytical pipeline. This study presents a generalizable, high-throughput pipeline that, for the first time, seamlessly integrates image acquisition in three steps, namely (i) automated droplet detection and cropping, (ii) quantitative texture analysis, and (iii) interpretable machine learning for disease classification. The label-free workflow operates with minimal user intervention, adaptable across imaging modalities and various biofluids, and experimental environments. Robustness of this pipeline is demonstrated on biofluids from both mice and humans, identifying disease-specific morphological signatures of influenza and diabetes directly from dried droplet patterns. As prior studies have not characterized dried droplet morphologies for these diseases, this work establishes a foundational benchmark for image-based diagnostic screening. By transforming endpoint morphologies into reproducible biomarkers, the pipeline enables rapid disease detection from blood dried patterns and is readily extendable to other biofluids, including saliva, sweat, and urine. Its simplicity, scalability, and cross-species applicability make it ideally suited for point-of-care diagnostics, particularly in resource-limited settings. To ensure reproducibility and facilitate broad adoption, open-access analytical tools with a graphical user interface for droplet region selection are provided, promoting translation to diagnostic, veterinary, and biotechnological applications.

Cite this article

Download citation ▾

Anusuya Pal, Piyush Thakur, Amalesh Gope, Ryuta Uraki, Kawaoka Yoshihiro, Miho Yanagisawa. Dried biofluid droplet morphologies for automated and scalable disease classification. Droplet, 2026, 5 (2) : e70054 DOI:10.1002/dro2.70054

登录浏览全文

4963

注册一个新账户忘记密码

References

Publishing order | Descend order by publishing year | Descend order by cited within

[1]	Pal A, Gope A, Sengupta A. Drying of bio-colloidal sessile droplets: advances, applications, and perspectives. Adv Colloid Interface Sci. 2023;314:102870.

[2]	Kim N, Li Z, Hurth C, Zenhausern F, Chang S-F, Attinger D. Identification of fluid and substrate chemistry based on automatic pattern recognition of stains. Anal Methods. 2012;4:50-57.

[3]	Pal A, Gope A, Obayemi JD, Iannacchione GS. Concentration-driven phase transition and self-assembly in drying droplets of diluting whole blood. Sci Rep. 2020;10:18908.

[4]	Yakhno TA, Yakhno VG, Sanin AG, et al. The informative-capacity phenomenon of drying drops. IEEE Eng Med Biol Mag. 2005;24:96-104.

[5]	Brutin D, Sobac B, Loquet B, Sampol J. Pattern formation in drying drops of blood. J Fluid Mech. 2011;667:85-95.

[6]	Mukhopadhyay M, Ray R, Ayushman M, et al. Interfacial energy driven distinctive pattern formation during the drying of blood droplets. J Colloid Interface Sci. 2020;573:307-316.

[7]	Pal A, Gope A, Iannacchione G. Temperature and concentration dependence of human whole blood and protein drying droplets. Biomolecules 2021;11:231.

[8]	Zeid WB, Vicente J, Brutin D. Influence of evaporation rate on cracks formation of a drying drop of whole blood. Colloids Surf. A. 2013;432:139-146.

[9]	Sobac B, Brutin D. Structural and evaporative evolutions in desiccating sessile drops of blood. Phys Rev E. 2011;84:011603.

[10]	Smith FR, Buntsma NC, Brutin D. Roughness influence on human blood drop spreading and splashing. Langmuir 2017;34:1143-1150.

[11]	Chen R, Zhang L, Shen W. Controlling the contact angle of biological sessile drops for study of their desiccated cracking patterns. J Mater Chem B. 2018;6:5867-5875.

[12]	Iqbal R, Shen AQ, Sen A. Understanding of the role of dilution on evaporative deposition patterns of blood droplets over hydrophilic and hydrophobic substrates. J Colloid Interface Sci. 2020;579:541-550.

[13]	Zurbriggen L, Baumgartner S, Schaub N, Kokornaczyk MO. Influence of temperature and relative humidity on patterns formed in dried plasma and serum droplets. Colloid Interface Sci Commun. 2022;49:100645.

[14]	Efstratiou M, Christy JR, Bonn D, Sefiane K. Transition from dendritic to cell-like crystalline structures in drying droplets of fetal bovine serum under the influence of temperature. Langmuir 2022;38:4321-4331.

[15]	Wang Y, Liu F, Yang Y, Xu L-P. Droplet evaporation-induced analyte concentration toward sensitive biosensing. Mater Chem Front. 2021;5:5639-5652.

[16]	Pal A, Gope A. Texture identification in liquid crystal-protein droplets using evaporative drying, generalized additive modeling, and K-means clustering. Eur Phys J E. 2024;47:35.

[17]	Batista BC, Tekle SD, Yan J, Dangi BB, Steinbock O. Chemical composition from photos: dried solution drops reveal a morphogenetic tree. Proc Natl Acad Sci. 2024;121:e2405963121.

[18]	Harindran A, Hashmi S, Madhurima V. Pattern formation of dried droplets of milk during different processes and classifying them using artificial neural networks. J Dispersion Sci Technol. 2022;43:1838-1847.

[19]	Acuña C, Mier y Terán A. Kokornaczyk MO, Baumgartner S, Castelán M. Deep learning applied to analyze patterns from evaporated droplets of Viscum album extracts. Sci Rep. 2022;12:15332.

[20]	Jeihanipour A, Lahann J. Deep-learning-assisted stratification of amyloid beta mutants using drying droplet patterns. Adv Mater. 2022;34:2110404.

[21]	Pal A, Yanagisawa M. Pattern recognition of drying lysozyme-glucose droplets using machine learning classifiers. Physica A. 2024;654:130141.

[22]	Vaez S, Dadfar B, Koenig M, Franzreb M, Lahann J. Deep learning-based classification of histone—DNA interactions using drying droplet patterns. Small Sci. 2024;4:2400252.

[23]	Pal A, Gope A, Yanagisawa M. From droplet to diagnosis: spatio-temporal pattern recognition in drying biofluids. Adv Intell Syst. 2025:e202500550.

[24]	Yang R, Cao N, Yang Y, et al. Label-free urinary protein detection through machine learning analysis of single droplet evaporation patterns. Anal Chim Acta. 2025;1356:344034.

[25]	Arévalo LA, O'Brien SA, Antonova O, Seifert A. Drying patterns of cerebrospinal fluid as indicator for Alzheimer's disease by a machine learning framework. J Phys Conf Ser. 2022;2407:012027.

[26]	Sefiane K, Duursma G, Arif A. Patterns from dried drops as a characterisation and healthcare diagnosis technique, potential and challenges: a review. Adv Colloid Interface Sci. 2021;298:102546.

[27]	Jocher G, Qiu J, Chaurasia A. Ultralytics YOLO. 2023. https://github.com/ultralytics/ultralytics

[28]	Van Griethuysen JJ, Fedorov A, Parmar C, et al. Computational radiomics system to decode the radiographic phenotype. Cancer Res. 2017;77:e104-e107.

[29]	Haralick RM, Shanmugam K. Dinstein IH. Textural features for image classification. IEEE Trans Systems Man Cybernetics. 2007:610-621.

[30]	Pedregosa F, Varoquaux G, Gramfort A, et al. Scikit-learn: machine learning in Python. J Mach Learn Res. 2011;12:2825-2830.

[31]	Pal A, Gope A, Athair AS, Iannacchione GS. A comparative study of the drying evolution and dried morphology of two globular proteins in de-ionized water solutions. RSC Adv. 2020;10:16906-16916.

[32]	Pal A, Gope A, Kafle R, Iannacchione GS. Phase separation of a nematic liquid crystal in the self-assembly of lysozyme in a drying aqueous solution drop. MRS Commun. 2019;9:150-158.

[33]	Pal A, Gope A, Iannacchione GS. A comparative study of the phase separation of a nematic liquid crystal in the self-assembling drying protein drops. MRS Adv. 2019;4:1309-1314.

[34]	Pal A, Gope A, Iannacchione GS. Hierarchical exploration of drying patterns formed in drops containing lysozyme, PBS, and liquid crystals. Processes. 2022;10:955.

[35]	Pal A, Gope A, Iannacchione GS. Statistical image analysis of drying bovine serum albumin droplets in phosphate buffered saline. In: Dash S, Pani SK, Balamurugan S, Abraham A, eds. Biomedical Data Mining for Information Retrieval, Scrivener Publishing LLC; 2021:213-235.

[36]	Pal A, Sengupta A, Yanagisawa M. Role of motility and nutrient availability in drying patterns of algal droplets. Sci Rep. 2024;14: 23481.

[37]	Sherwani SI, Khan HA, Ekhzaimy A, Masood A, Sakharkar MK. Significance of HbA1c test in diagnosis and prognosis of diabetic patients. Biomarker Insights. 2016;11:BMI-S38440.