Codic - Cough Count and Sound Monitoring With Pre-diagnosis

Cough is amongst the most common reasons for seeking medical service. In China, patients with cough account for more than a third of the total patient number in respiratory clinics, and it is estimated that more than 90 million individuals in China are suffering from this condition. While coughing can be an indicator of a serious disease (asthma, lung cancer), the vast majority of coughs can go away without medical treatment. Therefore, patients tend to ignore it until it starts getting worse. Cough misdiagnosis is common since many physicians have limited hearing resolution and don't have access to adequate historical data.

Cough is one of the most common symptoms of COVID-19. Our detection system can narrow down the suspects for COVID-19. If PCR-tests were required only for those suspects, we could end up saving money, resources, and time.

Two alternatives to signal retrieval:

• The app - utilization of the phone recorder

• A wearable sensor

Mel frequency cepstral coefficients:

• The algorithm uses a set of Mel triangle filters to extract the most relevant information from the frequency spectrum. The power of the signal that gets through each filter is the feature characterizing each cough.

• After recognizing and extracting the cough signal, the app counts the number of coughs. The app uses a linear regression algorithm to classify the cough into categories such as dry, wet, etc. This information can give users a pre-diagnosis.

The difficulties: Creating a huge database with all the cough patterns.

Available Problems:

• Privacy issue: devices collect users’ medical data by recording all day.

• Trust issue: users may cheat to avoid quarantine, so it’s hard for government and medical organizations to believe our devices’ results.

Response: Blockchain and TEE (Trusted Execution Environment)

• These encryption algorithms can protect our data.

• The immutable and decentralized system can offer trusted results.

• Our application is established to achieve Goal 3: Good health and well-being.

• Furthermore, in the pandemic context and in China (with strict COVID rules), this app works as a prediction app, which helps to reduce the time of testing and predicting.

• Award for Technical Innovation Team, Tsinghua Global Summer School 2022

• Hackathon 4: Future Information Technology for Human Healthcare