1. Can you please briefly summarize the paper? 

This study presents a diagnostic platform integrating field-effect transistor (FET) technology, a paper-based analytical device, and deep-learning (DL) based kinetic analysis to enhance quantitative biosensing performance. This platform demonstrated highly sensitive, rapid (< 1:30 min) and low-cost (< $0.15) detection of target biomarkers in human plasma. Deep learning is utilized to address critical issues such as sample matrix interference in FET biosensors by analyzing kinetic data from specific bioreactions. The effectiveness of this approach is validated through a proof-of-concept study, which demonstrates high accuracy in cholesterol testing?achieving a coefficient of variation below 6.46% and a correlation coefficient (r²) over 0.976 when compared to results from a CLIA-certified clinical lab instrument. This proof-of-concept signifies a breakthrough in biosensing, potentially revolutionizing point-of-care diagnostics and at-home testing by improving their accessibility, ease of use, and accuracy.

2. Can you please tell us the main difficulties you had in the laboratory work and how you overcame them?

While our DL algorithm was effective in enhancing cholesterol quantification within individual sets (with separate models trained for each of the three sets), it showed lower performance when training across different sets. A single model trained on data from all three batches demonstrated lower performance (r² < 0.9), highlighting the influence of additional factors on inter-batch repeatability. These factors include the varying properties of electrodes used in different testing batches, variations in enzyme and reagent concentrations due to handling issues, limited control over environmental factors such as temperature and humidity, variability between reagent batches, residual nonspecific binding of proteins in plasma on the electrode, and varying enzyme activity influenced by the pH or/and ion concentrations of plasma samples. These factors can be addressed in future iterations through quality controls implemented in the fabrication and assembly processes. Additionally, environmental factors that have a direct impact on the captured data can be added to the input of future inference models to improve the generalizability of the concentration inference model to different batches.

3. Please introduce your laboratory, university or organization to bio-researchers in Korea.

Junhong Chen Research Group at the University of Chicago primarily focuses on the molecular engineering of nanomaterials and nanodevices. We specialize in creating hybrid nanomaterials that feature rich interfaces, which are pivotal for advancing sustainable energy, environmental solutions, and biosensors. Our lab recently helped develop and publish the Water + AI strategy, which provides the framework for applying artificial intelligence to water-related problems. Junhong Chen is currently Crown Family Professor of Pritzker School of Molecular Engineering at the University of Chicago and Lead Water Strategist & Senior Scientist at Argonne National Laboratory. He also serves as the Science Leader for Argonne’s presence in the City of Chicago (Argonne in Chicago). ?Prior to coming to Chicago, Dr. Chen served as a program director for the Engineering Research Centers program of the US National Science Foundation (NSF) and the director of NSF Industry-University Cooperative Research Center (I/UCRC) on Water Equipment & Policy (WEP).

4. Please tell us your experiences and your thoughts related to research activities abroad.

Researching abroad offers an opportunity to immerse in a different cultural and academic environment. This exposure broadens one's perspective, enhances personal and professional growth, and fosters a deeper understanding of global scientific challenges and methodologies. Working in a foreign setting allows researchers to build international networks and collaborate with a diverse group of peers and experts. These relationships can lead to long-term partnerships, offering mutual benefits through shared resources, knowledge, and expertise.

5. Can you provide some advice for younger scientists who have plans to study abroad?

Investigate potential programs and institutions that align with your research interests. Being proactive is essential for young scientists planning to study abroad. Start by thoroughly researching potential programs and institutions well in advance, actively seeking scholarships, and preparing necessary documentation early to avoid last-minute pressures.

6. Future plan?

I will be joining Kompass Diagnostics as the co-founder and Chief Science Officer this November as it has recently closed $1.6 million seed funding. Kompass is building a pocket-sized diagnostic device to facilitate care delivery models beyond hospital walls. Kompass is leveraging the FET/paper-based diagnostic platform to develop a highly cost-effective, lab-grade blood diagnostics that can deployed in any clinical environment. 

7. Do you have anything else that you would like to tell Korean scientists and students?

Engaging in translational research offers a valuable opportunity to gently bridge the gap between laboratory discoveries and their practical applications. This approach helps to subtly guide scientific findings from the bench to broader real-world impacts, particularly in sectors like healthcare and technology. By focusing on translational research, you can contribute thoughtfully to advancing innovations that meet societal needs while fostering collaboration between academia and industry.