Sandeep Choudhary
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Sandeep Choudhary has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
Abhijeet Joshi
Indian Institute of Technology, Indore
3 shared publications- Abhijeet Joshi
Indian Institute of Technology Indore Khandwa Road
3 shared publications Tanmay Vyas
Department of Biosciences and Biomedical Engineering, Indian Institute of Technology Indore, Khandwa Road, Indore, 453552, Madhya Pradesh, India
2 shared publicationsBhavana Joshi
1 shared publicationGaurav Pandey
1 shared publication- Gaurav Pandey
Department of Biological Sciences & Bioengineering, Indian Institute of Technology Kanpur, Kalyanpur, UP, 208016, India
1 shared publication
Latest Publications
- 30 July 2026
- Published by Taylor & Francis
- p. 89-106
Machine Learning for Medical Data Classification
Supervised learning, especially machine learning (ML), has proven to be a valuable and powerful tool in biomedical data classification. ML algorithms operate by training a classifier using labelled data, by learning its pattern, the trained model makes predictions based on unlabelled and unknown data. In recent years, ML has gain prominence in the area of biomedical engineering, particularly for its use in the classification and interpretation of intricate medical data, including imaging, genetic information, and physiological signals. These algorithms are assistive technologies that aid clinicians by offering data-driven insights, which can be particularly useful in time-critical situations like medical emergencies. In a standard ML pipeline, features of interest are initially extracted from the raw biomedical data. These features are subsequently employed to train a classifier—a model that learns to differentiate between various categories or outcomes. After training, the classifier can then be used to classify unseen or novel data, allowing it to make predictions or aid in diagnosis with a level of accuracy and reliability that complements human skill. The curse of dimensionality is an issue in ML algorithms; this chapter begins with a discussion on the reduction of dimension, such as Principal Component Analysis (PCA). ML classification algorithms such as support vector machine (SVM), K-Nearest Neighbour (KNN), classification trees, random forest, and adaptive boosting algorithms are discussed with case studies.
- 30 July 2026
- Published by Taylor & Francis
- p. 89-106
Machine Learning for Medical Data Classification
Supervised learning, especially machine learning (ML), has proven to be a valuable and powerful tool in biomedical data classification. ML algorithms operate by training a classifier using labelled data, by learning its pattern, the trained model makes predictions based on unlabelled and unknown data. In recent years, ML has gain prominence in the area of biomedical engineering, particularly for its use in the classification and interpretation of intricate medical data, including imaging, genetic information, and physiological signals. These algorithms are assistive technologies that aid clinicians by offering data-driven insights, which can be particularly useful in time-critical situations like medical emergencies. In a standard ML pipeline, features of interest are initially extracted from the raw biomedical data. These features are subsequently employed to train a classifier—a model that learns to differentiate between various categories or outcomes. After training, the classifier can then be used to classify unseen or novel data, allowing it to make predictions or aid in diagnosis with a level of accuracy and reliability that complements human skill. The curse of dimensionality is an issue in ML algorithms; this chapter begins with a discussion on the reduction of dimension, such as Principal Component Analysis (PCA). ML classification algorithms such as support vector machine (SVM), K-Nearest Neighbour (KNN), classification trees, random forest, and adaptive boosting algorithms are discussed with case studies.
- 24 April 2022
- Published by The Electrochemical Society in ECS Transactions
- Vol. 107 (1) , 8495-8504
Fluorescence-Based Sensing Assay for Point of Care Detection of Healthcare Parameters in Food Samples
Nowadays, awareness and concern regarding food safety are growing publicly in developing countries across the world. Various factors like sickness, misery, and even death are caused by contaminated food consumption. The increasing population across the globe increases food consumption. To overcome the gaps between the demand and supply of food products, people practice adulterating the food products (urea in milk). This work reports the synthesis of a solution-phase system containing fluorophores (FITC: act as indicator and Rubpy, serves as a reference) and enzymes (urease) for the real-time monitoring of samples. A color sensor-based device (CSD) has been developed and used to analyze the pH (4-9 pH) and urea (0-70 mM) in milk. The results obtained on CSD were cross-referenced on FOS simultaneously and found to be suitable for detecting degradation and urea adulteration in milk.
- 24 April 2022
- Published by The Electrochemical Society in ECS Transactions
- Vol. 107 (1) , 8495-8504
Fluorescence-Based Sensing Assay for Point of Care Detection of Healthcare Parameters in Food Samples
Nowadays, awareness and concern regarding food safety are growing publicly in developing countries across the world. Various factors like sickness, misery, and even death are caused by contaminated food consumption. The increasing population across the globe increases food consumption. To overcome the gaps between the demand and supply of food products, people practice adulterating the food products (urea in milk). This work reports the synthesis of a solution-phase system containing fluorophores (FITC: act as indicator and Rubpy, serves as a reference) and enzymes (urease) for the real-time monitoring of samples. A color sensor-based device (CSD) has been developed and used to analyze the pH (4-9 pH) and urea (0-70 mM) in milk. The results obtained on CSD were cross-referenced on FOS simultaneously and found to be suitable for detecting degradation and urea adulteration in milk.
- 1 April 2022
- Published by Elsevier in International Dairy Journal
- Vol. 127
Development of an embedded system for real-time milk spoilage monitoring and adulteration detection
- 1 April 2022
- Published by Elsevier in International Dairy Journal
- Vol. 127
Development of an embedded system for real-time milk spoilage monitoring and adulteration detection
- Open Access
- 2 July 2020
- Published by Springer Nature in Scientific Reports
- Vol. 10 (1) , 1-11
Ultrasonic atomizer based development of pH sensor for real time analysis
Fluorescent pH biosensors have gained importance owing to their low cost utilization in real time monitoring of biological and food samples in comparison to conventional pH meters. The research reports a novel method of ultrasonic atomization for developing a fluorescent pH sensor for real-time analysis made of Fluorescein isothiocyanate (FITC)-dextran/FITC-dextran-Tris (2, 2′-bipyridyl) dichlororuthenium (II) hexahydrate as indicator and reference fluorophores, respectively. The process of ultrasonic atomization ensures formation of monodisperse dye immobilized alginate microspheres ensuring efficient pH sensing. The developed biosensor was tested on milk samples, which has a short life span and shows a significant fall in pH with time due to microbial spoilage. The proposed biosensor showed a linear range of pH 4–8 (R2 between 0.96–0.99 for different single/dual fluorophore biosensors) which suitably cover the pH of milk during the entire storage period and spoilage. The % recovery for predicted pH falls between 90–110% compared against standard pH meter, indicating a good accuracy of estimation and low turnaround time (10 min). Thus, real-time monitoring using fluorescent pH biosensor for milk samples may profoundly improve the economics of losses occurring in processing and storage with capability of in-package continuous quality assessment.
- 2 July 2020
- Published by Springer Nature in Scientific Reports
- Vol. 10 (1) , 1-11
Open AccessUltrasonic atomizer based development of pH sensor for real time analysis
Fluorescent pH biosensors have gained importance owing to their low cost utilization in real time monitoring of biological and food samples in comparison to conventional pH meters. The research reports a novel method of ultrasonic atomization for developing a fluorescent pH sensor for real-time analysis made of Fluorescein isothiocyanate (FITC)-dextran/FITC-dextran-Tris (2, 2′-bipyridyl) dichlororuthenium (II) hexahydrate as indicator and reference fluorophores, respectively. The process of ultrasonic atomization ensures formation of monodisperse dye immobilized alginate microspheres ensuring efficient pH sensing. The developed biosensor was tested on milk samples, which has a short life span and shows a significant fall in pH with time due to microbial spoilage. The proposed biosensor showed a linear range of pH 4–8 (R2 between 0.96–0.99 for different single/dual fluorophore biosensors) which suitably cover the pH of milk during the entire storage period and spoilage. The % recovery for predicted pH falls between 90–110% compared against standard pH meter, indicating a good accuracy of estimation and low turnaround time (10 min). Thus, real-time monitoring using fluorescent pH biosensor for milk samples may profoundly improve the economics of losses occurring in processing and storage with capability of in-package continuous quality assessment.