Drug Supply Chains: How Blockchain and ML can be Enablers

This piece refers to the article –

https://www.supplychainbrain.com/blogs/1-think-tank/post/31947-applying-blockchain-and-machine-learning-to-drug-supply-chains

Since long, pharma companies have been unable to track products through their respective supply chains. This lacuna has made it easier for counterfeit drugs to slip into the drug supply chains for some time now.

To counter this challenge, new track-and-trace systems are needed. Experts believe that blockchains can be used as panaceas in this case. Blockchains can actually provide tech base upon which it would be easier to track legit drugs and avoid/mitigate the instances of counterfeits to circulate in the market freely.

The article mentioned and I quote –

“Counterfeit drugs are defined by the World Health Organization as those that “are manufactured fraudulently, mislabeled, [of] low quality, hiding the source detail or identity, and [not following] the defined standard.””

Now, counterfeit drugs are not totally fake. They may contain some genuine medicinal ingredients, but they also do contain certain toxic substances at the time of their production and can cause serious health issues post-consumption. The article further mentioned, and I quote –

“Sometimes the makers of counterfeit medications use the logos of reputable companies to get their products into the market. Although a worldwide problem, this practice disproportionately affects developing countries.”

These non-legit drugs are distributed vide a complex network, thus, making them difficult to nip in the bud. Preventing their distribution must be the order of the day and this can only be possible via a system which enables superior track and trace capabilities. Blockchains fit the bill well here.

Blockchains would offer an electronic ledger which would enable all the participants in the network to view and validate the information, esp. with respect to application of blockchains in healthcare/drug supply chains.

The article further mentioned, and I quote –

“To create a secure drug supply-chain management system, designers proposed the use of blockchain with Hyperledger fabric. This software is capable of monitoring and tracking all parts of the drug-delivery process. It can configure multiple world state databases to maintain a set of current values. When applied to the pharmaceutical world, it enables medicines to be accurately traced regardless of where they are in the world.”

This presents itself as a major USP and a differentiator when put to use in drug supply chains. Experts and researchers propose a Blockchain and ML based “Drug Supply chain management and Recommendation System (DSCMR)”. The article further mentioned, and I quote –

“It’s important to distinguish between these two complementary components — in the drug-management system, users can track the drug at every step, make orders, update orders, and more, while the recommendation system identifies the best medicine for pharmaceutical customers through the use of N-gram and LightGBM machine learning-based modules.”

The Drug supply chain management system would enable users to perform transactions such as check drug information; track and trace orders; and update records. Users may include patients, doctors, manufacturers, distributors, pharmacies, and hospitals. Now, the data pertaining to the relevant participants in the chain is stored in the drug supply chain blockchain. In addition, each of the participants (users) is provided with a webapp which would enable them to perform transactions and communicate with the blockchain network, in addition to tracking delivery statuses. Now, to ensure that security remains intact all through, the users receive permission to check the details of the drugs in entirety, using client app to login and perform the respective transactions.

The article further mentioned, on the actual process flow as well and I quote –

“To submit a transaction proposal, users must make a request using their credentials. If deemed valid, the request is sent to peer nodes in the system to be reviewed and approved. These nodes examine the proposal and grant approval if it fulfills the smart contract criteria. They also validate the results of the transactions and record them in the ledger. Once this action has been completed, the ledger updates all data for everyone to see.”

The drug-recommendation system, on the other hand, uses a public review dataset. This public review dataset is based on reviews of drug users and was created to train the system and then predict the most effective drugs available industrywide. The review data also includes information on side effects; benefits; and customer comments received in the client app.

The recommendation system self-trains itself through the N-gram model as and when it recognizes latest reviews and thus, updates the results accordingly.

Further, the use of Smart Contracts limits the number of users involved in a specific transaction.

The article further also mentioned, and I quote –

“Several conclusions emerged from testing the proposed machine learning-based DSCMR system. In the drug-management system, the client application mentioned earlier can communicate with the blockchain effectively. Once a user’s identity is validated by the peer nodes, they are able to start transactions….. manufacturers can add, update, or delete drug details in the blockchain. Other participants such as doctors, hospitals, and pharmacies can view these records and make changes and updates to the drug information as needed”

The DSCMR system offers 4 major advantages –

·        Recommend top-rated medicines to customers

·        Allow patients to access the system through a client app and customers (doctors) using their credentials

·        Provide customers with a secure and transparent experience

·        Patients can verify the authenticity of their drugs by a scan of the barcode

Blockchains would surely end up delivering and unlocking greater value esp. in case of drug supply chains globally. This would be more beneficial to drug supply chains than perceived or thought previously. When it is the matter of life or death as it is in this case, using a technology which can relied upon is a more prudent approach rather than leaving things to chance and hoping that the drug being administered to a patient or a patient taking a drug is legit enough and would be effective as expected… With Blockchains and ML, patients and providers alike can be assured that only legit drugs are administered at all times….