LifeLit: Bacteria Detector for Resource Limited Settings
July 2014 - September 2014
Stanford Mechanical Engineering Research Lab(MERL)
In Summer 2014, I was selected as one of the 18 undergraduate visiting researchers to do independent research sponsored by School of Engineering at Stanford University (UGVR Program). Under supervision of Prof. Sindy Tang, I designed a low-cost chip for disease detection at resource limited areas. The work was presented at Stanford UGVR research symposium and published at Biomicrofluidics in 2015 (Link).
Motivate
-
Tuberculosis(TB) is one of the most deadly diseases that kills over one million people each year and infects one-third of the world's population.
-
Early diagnosis of TB is crucial to the prevention and control of disease. Standard diagnostic methods takes at least 48 hours, which allows the spread of the disease.
-
A rapid, accurate, and automatic method with low-cost equipment for detecting mycobacterium tuberculosis (Mtb, the cause agent of TB) is needed, and the concentration of it in drinking water reservoir needs to be real-time detected.
Ideate
-
Disease test is like a catching-and-counting bacteria game.
-
Usually the concentration of bacteria in the sample is too low to detect.
-
Our idea is to break the sample into millions of droplets.
While it's hard to catch a fish in a sea, isn't it much easier to catch it in a bucket?
Early sketch of ideas
Concept skecth
Develop
-
Created an optical system to activate and capture fluorescence.
-
Fabricated microfluidic chips to generate droplets.
-
Developed a software interface as automatic monitor.
Marker and probe for bacteria detection
-
We use BlaC (an enzyme naturally expressed by tubercle bacilli) as a marker and a designed BlaC-specific fluorogenic substrates as probes for Mtb detection.
-
In the presence of bacteria, specific enzymes in the bacteria convert the probe substrate into a fluorescent product.
Schematic description of droplet-based device
-
Monodispersed droplets containing bacteria and probe are generated via a flow focusing device, and incubated for fluorescence to turn on.
-
Droplets are reinjected into a narrow constriction and fluorescent droplets are detected by photomultiphier and counted by software. The ratio of fluorescent droplets corresponds to the concentration of bacteria in the sample.
Experimental setup and software interface
-
Droplets with bacteria are excited by blue light and will let off green light, which is detected by a photomultiphier.
-
With a Labview DAQ, signal of light intensity can be recoreded at a sample rate of 125,000, and automatically processed in the software.
Test
-
Bacteria with concentration of 500 cell/mL has been successfully detected within 4 hours.
-
The detection limitation is lowered by 1000 times.
Detection of Mtb with ultra-low concentration
-
Signal of fluorescein and empty droplets mixture flowing through a constriction.
-
Concentration calculated from droplet-based method and real concentration.
Deliver
-
Publication: "Quantitative detection of cells expressing BlaC using droplet-based microfluidics for use in the diagnosis of tuberculosis", Biomicrofluidics, 9, 044120, 2015.
-
Presentation: Stanford UGVR research symposium. Click here to download the presentation powerpoint.
Skill Sets
Signal Processing
Optical Design
Chip Design
Labview Programming
Matlab
Autocad/Solidworks
MEMS/IC/Microfluidics Fabrication
Medical Device Deisgn
