Team:Patras/Proposed Solution

Home

Proposed Solution

- Portable molecular biology laboratory

BentoLab, the Portable Molecular Biology Laboratory, is a fully equipped 3.5kg portable laboratory with dimensions of 330mm*214mm*81mm, making it easy to carry in a suitcase. It has been introduced as a hardware device by the UCL_E team at the iGEM 2013 competition. It contains all the needed instruments to accomplish a genomic analysis, but by now, it has only been used before for educational purposes.

The BentoLab includes a 32-sample capacity thermal cycler, with a dynamic temperature range of 12-105 ° C, a ramp rate of 2.5 ° C/sec, and temperature accuracy of ± 0.2 °C of the programmed target temperature. It features a microcentrifuge of 6x1.5mL microcentrifuge tubes (known as Eppendorf tubes) capable of rotation up to 13,500 rpm (8,000xg) and automatic control of the safe closure.

Mobirise

Besides, the portable lab has a 50-120V electrophoresis device with a 91mmx78mm gel disc, nine and twelve-position combs, and a built-in power supply. The electrodes are made of platinum for proper operation and greater durability. Finally, the gel tank is molded to prevent leaks with a vented lid to dissipate heat during electrophoresis.

Finally, it has a transient electronics Design of Blue LED array (468 nm) with a surface area of 105mm x 85mm made of scratch-resistant acrylic plastic. Finally, a built-in plastic rack of 18 positions for 1.5-2 ml microtubes and 32 positions for 0.5 ml makes the operators work easier. Simultaneously, a Wi-Fi connection is available for easy software updates, and its power supply is designed for maximum flexibility thanks to the built-in power adapter.

Our project aims to use this portable laboratory to perform a fast genome analysis assisted by an Artificial Intelligence algorithm.

- Artificial Intelligence

In our project, we use AI to automate a part of the procedure to be easily understood by everyone. At first, to achieve that, we applied machine learning and computer vision techniques and created a user-friendly system where a health professional can use it. Electrophoresis gel images from the BentoLab are used to train deep learning object detection models and test whether they correctly classify a patient’s genotype to determine the exact administered dosage. Our models’ validity is proved using a set of test images that haven’t been used during the training process and images uploaded to our system by the three medical students we invited to test our BentoLab-AI system.

Mobirise

Our task was to locate and classify different objects of interest (OoI) in electrophoresis gel images. Each image consists of six (6) sample columns, each consecutive pair representing a person’s sample, two (2) blank columns, and the ladder. To perform multiple object detection, a modification of the state-of-the-art algorithm Yolo is implemented. A dataset is provided by BentoLab that relates the presence of a PCR product - fragment in each column to the genotype of the corresponding patient’s sample. Our team trained an algorithm to identify a PCR product and decide if someone is homozygous or heterozygous to the SLCO1B1 gene. The dataset contains twenty (20) images, and there are four (4) possible combinations of genes for each column pair.

Project Hippocrates proves to be very useful, assisting physicians by proposing the preferred drug dosage while prescribing a drug. What makes it unique is the provided step-by-step guidance and the AI algorithm that is being used to translate the genetic data into useful clinical information.

- What is the potential of our MiniLab – AI system?

- What about the cost?

The operating cost of genetic analysis in our portable molecular biology laboratory is low compared to the cost of performing the same analysis in conventional laboratory devices. Thus, professionals and patients could save money from not transporting blood samples in laboratories based in big metropolitan areas.

-What if experts from the laboratory were offered to move to the patient’s place in order to take his/her sample back to the laboratory for examination? What about the time needed for a genetic test to be performed in a conventional lab?

Through our system, the time needed for genomic analysis is decreased from 5-10 days to only 5-6 hours compared to existing methods. This can be characterized as one of the most important advantages as a physician could have the results only in a few hours to prevent his patient’s side effects and avoid hospitalization.

- What if we build new laboratories in some of the inaccessible, remote areas and these laboratories scientists could examine local peoples’ samples?

The cost of building a new laboratory in an inaccessible area is higher than purchasing our BentoLab-AI system. Then, the cost of performing a genomic analysis is ten-times lower than in a conventional laboratory. Thus, we believe that it is more beneficial for a physician and the patients to use our proposed system when traveling to a remote area than establishing a new one in this place.

- What about the direct-to-consumer companies’ policy, where the patients send the sample from their home to the lab?

Patients may not be capable of collecting their sample (saliva or blood) on their own, and they might not follow the right - difficult in handling – procedure. Our portable lab’s sample collection process is easier and more efficient, while a few requirements are requested for the test operation and storage.

- What about the accuracy and the utility of the results?

This testing device provides analytical validity (presence/absence of a genetic variant), clinical validity (relation of the genetic variant to the presence/absence/risk of the disease), and clinical utility (the tests provide information about the treatment of the disease). By using our BentoLab-AI system, statin nonadherence and drug discontinuation are avoided.

- What about the direct-to-consumer companies’ policy, where the patients send the sample from their home to the lab?

Patients may not be capable of collecting their sample (saliva or blood) on their own, and they might not follow the right - difficult in handling – procedure. Our portable lab’s sample collection process is easier and more efficient, while a few requirements are requested for the test operation and storage.

- What about a patient’s psychology?

Using the BentoLab-AI system for genetic analysis, anxiety and other emotional risks are avoided due to the short time required to show the result. Thus, there is a lower impact on the patient’s occupation or personal life and risk for his/her life insurance to expire.

- Is training and familiarity with the field of Pharmacogenomics easy to be achieved with BentoLab?

Pharmacogenomics is a new and emergent track in the world of medicine. Thus, many physicians aren’t adequately qualified and need to be educated to use it in clinical practice. Nowadays, genetic tests are not widely used due to the lack of doctors’ education. For this reason, we first created step-by-step guidance to assist and encourage physicians to perform genetic analysis on their own.

- What about Artificial Intelligence innovation?

Combining AI with Pharmacogenomics in a diagnostic tool could have a significant impact on our society. Our AI system could operate numerous data that can be analyzed to find the right dosage for each patient. Although an AI system must be trained, it is less likely to make mistakes compared to human handling. Moreover, an average person will work for 4–6 hours a day, excluding the breaks. But using AI, we can make machines work 24x7 at a reduced cost, without any breaks, and they don’t even get bored, unlike humans. Using AI, we can make machines make decisions and carry out actions faster than humans. Artificial intelligence has massive potential advantages, as it could allow physicians to make decisions recommended by our BentoLab-AI system.

About

16 students from Patras blending Pharmacogenomics with Artificial Intelligence to redefine medicine

Follow Us