Sweet Sleep

A complete sleep monitoring package that precisely monitors the four sleep-wake stages, calculates the total sleep time, deduces the quality of sleep, and helps diagnose the sleep disorders if any.

I am highly interested to put forward a step towards the indigenous development of a low-cost wrist-worn sleep monitoring device that can precisely monitor the 4 sleep-wake stages including Wake, REM, light and deep sleep, total sleep time, quality of sleep, and will help diagnose the sleep disorders if any. My project is aimed to develop a system that contributes to medical technology by carrying out sleep research on the Pakistani adult population. The proposed device may satisfy the need for home-based, inexpensive, comfortable, and convenient sleep monitoring. The suggested system will allow people to play an effective role in maintaining their physical and mental health and will permit them to experience healthier sleep. Also, during this period of the pandemic, where it becomes difficult to go out for regular check-ups, it becomes important to permit people to age in place for as long as conceivable. 

The health and living standard of millions of people is adversely being affected by various sleep disorders which are proven to be health hazardous for up to 45% of people worldwide. It becomes necessary to analyze sleep, sleep cycles and diagnose sleep disorders at the earliest. Although many gadgets have been designed in this regard and their use is valued all over the world, very few sleep-related studies have been reported and there is limited availability and adoption of wearable sleep technology in Pakistan. Limited research reported in the literature indicates that more than 30% of the general adult population of Pakistan is suffering from insomnia and a major proportion is at high risk of obstructive sleep apnea. People with financial crises are 59% more likely to be affected by sleep disorders. 

The first problem that triggered the initiation of this project was the limited use of sleep monitoring devices and the lack of awareness about sleep-related disorders in Pakistan. There isn't any company that can locally manufacture such wearables and available devices are too expensive to purchase. Sleep patterns vary geographically, there isn’t any device that is trained and tested on Pakistani People.

According to a survey in Pakistan from 2003 to 2012, 38% of accidents on motorways and highways were due to sleep-deprived drivers. Out of which 54% of accidents caused on-the-spot death and 41% caused serious injuries. Sleeping on the wheel is something that could happen to the best of us. This is a major problem.

The target audience whom I would like to serve includes Adults (especially students) interested in self-care, doctors or sleep physicians who want to recognize their patients sleep patterns, older adults on bed rest (the device provides data regarding mobility, their caretakers can be alerted to make them move to keep them safe from bedsores), drivers driving for long hours, people working in construction areas, paramedics and employees working in night shifts. The device is specifically advantageous for sleep researchers to get data to work on. 

Through this project, I want to target sustainable development goal number three that is “Health and well-being”. Particularly, I want to promote mental health as poor sleep triggers viral loneliness and social rejection, causing a variety of mental health problems. This will be done using the sleep monitoring capability of the device. I want to put efforts to reduce deaths and injuries due to road accidents through the driving mode, and I also want to achieve universal health coverage by the use of a smart alarm feature. The per-minute raw data will be provided to sleep researchers so to carry further studies in sleep medicine. The availability of raw data can help sleep researchers to explore the prevailing sleep disorders and to suggest treatments accordingly.

The global population is increasing and is expected to increase over the next forty years, causing an immense load on health care solutions. This has led to the home health monitoring systems. Goals of national and personal development are achieved through a healthy population. Sleep is one of the five basic survival needs of the human being. The overall health status of an individual may depend on multiple factors including physical wellbeing, mental health, social comfort, and emotional fitness and sleep is directly or indirectly related to all these factors. 

 I observed sleep disorders having their deep roots in society especially in students and older adults. I researched sleep medicine and the state of the art and looking at the severity of sleep-associated problems I found that it is important to immediately put forward a step towards the indigenous development of a low-cost sleep monitoring device that can precisely monitor sleep and will be easily accessible to the majority of the population. Such a system will be extremely helpful for individuals interested in self-care and for sleep physicians who desire to regularly monitor the sleep habits of the patient for diagnosis and cure purposes. 

Yes! I have engaged potential users in the design and development of this prototype. The device is being trained and tested on the Pakistani population; due to this reason, it can recognize the local’s sleep pattern easily.

Up till yet, only the working version of the device is ready. The device still needs to be validated against Polysomnography, a gold standard reference in sleep monitoring. after validation, the device can be commercialized for sale in the market and for use by the locals. However, we have tested the machine learning model on which the device is working and the results are very promising. 

We have trained the model on 9 people and tested it on 6 persons yet. Our target is to at least test our device on 40 people of different age groups so as to get varying sleep patterns associated with the people of Pakistan. The data set consisted of 5000 + entries. The confusion matrix indicated that the model is about 83 % accurate, we have got an average precision-recall of 91%. The area under the curve (AUC) of the micro-average ROC curve came out to be 0.97 and that of the macro-average ROC curve was 0.96, AUC of all the stages predicted was greater than 0.90. The cohen’s kappa score of the predicting model was 0.682 which indicated substantial agreement of the predicted stages.

The solution is mainly based on an IMU sensing technology, the sensor receives 9 DOF data frpm patient body to classify it into sleep stages. A machine learning algorithm (Random Forest) is running behind the scene to analyse the data. the results are visualized on mobile phone. An interactive mobile appication developed to fullfill this prupose. So on whole we can say that the package is full of technology ranging from basic electronics to mobile application to machine intelligence. 

Our product is still in the prototype stage. We have not launched our product yet, but we are willing to serve atleast 1000 people all over the Pakistan, next year. Definitely, we intend to intensify this number in coming years.  

Through this project, I want to target sustainable development goal number three that is “Health and well-being”. Particularly, I want to promote mental health as poor sleep triggers viral loneliness and social rejection, causing variety of mental health problems. This will be done using sleep monitoring capability of the device. I want to put efforts to reduce deaths and injuries due to road accidents through the driving mode, and I also want to achieve universal health coverage by the use of smart alarm feature of my device. 

Progress measuring tools: 

The first evaluating tool is the accuracy I am getting from the developed prototype. 84% accuracy of the model shows that the sensor I am using can be used for sleep monitoring.

The second evaluating measure is the interest of people which they show when I advertise my device. This indicates that I am heading in the right direction, and I am able to solve a measure injustice which people suffer these days. This also include the number of views, likes, shares and subscribers of my video explaining the proposed system.

The third evaluating tool is the expense of my device. My device is perhaps a low-cost alternative and will be easily purchased by people. 

The forth and the most important tool to measure my success is the appreciation from entrepreneurs.

The only barrier I see at the moment is that my device needs to get a product shape. It needs to get validated so that I can make efforts for its legalization and commercialization. 

We are three members of the team. I am Andleeb Khowaja, a Mechatronics engineer. I am leading this project. The other members are Aamir Khowaja, a Mechanical engineer by profession who is handling the finance area; and Ali Aan Khowaja, who has developed the mobile application and website and is managing the software side of the project. Our team is supervised by Prof. Dr. Jawaid Daudpoto, Chairman Mechatronics Engineering department MUET, Jamshoro.  

No! we are not working with any team or organization currently.

I am not aware of the award that is mentioned. 

I am not a US resident.