Book a demo
Health, disease and aging affect everyone. Everyone invests in their well-being, for prevention and treatment. Investments are based on data, which are collected at home, in hospitals, clinical research laboratories and mobile devices. For security and privacy reasons, the data is kept separately.
Federated learning offers the possibility to address these issues, both for the general population with features common to all and for the individual in the rare or unusual case.
The variability of applications and configurations is another strong point of Federated Learning offered by Sensoworks and Frontiere:
Hospitals could use federated learning models to predict the likelihood of an individual patient developing a disease or contracting an infection after being admitted. This could help doctors make better decisions about how best to treat each patient based on their specific needs. Additionally, these predictions could be used by other organizations that work with hospital patients (such as insurance companies) in order to provide better coverage options for those who need it most.
Another area where federated learning can benefit healthcare is by helping us understand diseases like cancer or Alzheimer's disease at the genetic level better. Using federated learning models on DNA samples collected from thousands of people suffering from these conditions, researchers can gain insight into which factors contribute most to the development of these diseases.
Federated learning provides interoperability gains that positively impact healthcare professionals quickly accessing a patient’s medical information even if they are not within the organization’s internal systems (e.g., patients visiting from another country), and applying analytics. High quality medical decisions could be ensured regardless of the patient's treatment location and local knowledge of the disease. Physicians can augment their expertise with expert knowledge from other institutions, ensuring consistency of diagnoses that would not be possible otherwise, improving the effectiveness of patient care and enabling greater collaboration in the healthcare sector.
Federated learning has been used in medical research because it allows researchers to create large sets of patient data that they can then use as part of their research efforts without needing direct access to any individual's private health information. It also allows researchers who may not have access to certain medical records or resources (such as patients' genetic information) to instead access those same resources through other researchers' databases. This prevents privacy breaches while still allowing researchers access to sufficient information and insights.
As healthcare and clinical-data is scattered, gathering a data set that is complete enough to track rare cases involves combining data from various data silos. The practical challenges of any single site lacking adequate and sufficient data for rare adverse drug reaction detection and prediction, can be addressed. Likewise federated learning can assist in reaching the cohort sizes for orphan/rare disease studies and ensuring ethnic genomic diversity.
Precision medicine is a growing field in healthcare. A precision medicine study is an experimental study that uses personal data to develop a personalized treatment plan for an individual patient. The aim of these studies is to direct the treatment based on an individual person's unique genetic profile and medical history. Federated learning can be used to predict things like disease progression or patient outcomes, but it can also be used to predict patient responses to medications or other treatments. Using federated learning to analyze large amounts of patient data could help researchers identify new approaches to treating diseases like cancer and Alzheimer's at the molecular level, which could lead to more personalized treatments for patients. With access to large data sets, more precise drugs can be developed in an accelerated way.
Home health monitoring has attracted great attention for aging populations around the world. With abundant user health data being accessed by Internet of Things (IoT) devices, smart healthcare has seen many success stories. A personalized wearable device is used to collect, store and send health-related metrics, and can make medical recommendations for unusual health conditions. Nursing homes can offer specialized and constant care to the elderly. In these centers, remote health monitoring and recommendations can offer better services at reduced costs.
By unifying large data sets, new levels of well-being are offered at affordable costs.
Contact: dr. Remco Foppen
Fill out the form at the end of the page to request information on products or partnerships.
Data drives discovery. As a consequence one would think that more data would help drive more discovery, or at least accelerate discovery. To access more data oftentimes, one has to aggregate and centralize data from different sources. These sources can be from a variety of devices, instruments, formats, languages, countries and legislations. In the most straightforward approach, data is brought together and (pre-)processed at a single site and analysed centrally at the location of the AI model. This is ‘traditional’ centralized learning.
In practice, data is trapped. Data could be trapped in paper documents, pdfs, or in physical libraries. Intelligent Document Processing is a way to approach this type of unstructured ‘trapped’ data. Alternatively, data can be trapped by ownership, and privacy. As an example, an industrial partner may have an interest in a dataset, and therefore the owner of the data will be less willing to share the data. A patient, or an institution representing the patient, might be restricted to sharing the data for privacy reasons. A technical reason for data being trapped is when a device was developed without the objective of data aggregation, and thus lacking means for data harmonization altogether.
Driven by the advances in the Internet of Things, federated learning (FL) offers a solution, to release data and drive discovery. In FL, the model is trained without seeing or touching the data. Metaphorically speaking, the AI model travels to each dataset rather than data being aggregated at the site of the AI model. FL stimulates data sharing collaborations across sites without compromising patient privacy data legislation, governed by e.g. GDPR and HIPAA. In parallel, FL removes obstacles posed by heterogeneity of local devices or local (clinical) data management systems.
An example in biomedical research: In many instances the number of cases for orphan and rare diseases per institution are too low to garner the benefits of AI assisted prediction capabilities. Here FL trains a machine learning algorithm across multiple decentralized devices or servers, without actually exchanging the data or sharing any sensitive patient data. The federated approach promotes access to maximal data, while lowering institutional burdens to data sharing. Deploying FL over multiple sites and continents, prompted recent breakthroughs in the diagnosis of rare and aggressive cancers, like Triple-negative breast cancer and Glioblastoma. Thereby overcoming the lack of sufficient data for an AI analysis at a single site and thus outperforming the locally trained AI models.
Sensoworks, partner in the Frontiere network of companies, is in the initial phase of developing a FL model in collaboration with the Politecnico di Milano. Different FL architectures and types will be investigated to find the most optimal way to deploy an anomaly detection model on data derived from sensors, for which the benefits of FL outweigh the strenuous efforts of harmonizing data.
Contact: d.r. Remco Foppen
On the one hand, networked medical devices are revolutionising the way patients engage with healthcare. On the other hand, these devices are exposing millions of patients and healthcare providers to safety and security risks.
While in the past the healthcare industry focused largely on patient privacy, it is important to note that security is not the same as privacy. Privacy focuses more on access control, while security is about protecting the systems and sensitive data from intruders.
Many healthcare records and devices in use today are vulnerable, and the number of networked medical devices is rapidly growing. As more connected devices come to the market, the following security risks must be addressed.
Numerous industry working groups have convened to discuss and create standards, guidelines, and best practices for securing networked medical devices across the healthcare industry. This work is essential and will move the industry in the right direction, but there are security decisions that can be made today to protect devices and the sensitive data they transmit. Certain security approaches, such as encryption and authentication, will be part of whatever standard emerges from the working groups.
Secure IoT medical devices should contain costs for both industry professionals and patients. Today, we have much more data for medical research and patient monitoring to implement more effective risk prevention plans - in this context, better device safety means better healthcare.
Sensoworks IIoT Platform allows you to acquire data from any medical device - whether it is a “wearable” device such as a glucometer, or a diagnostic device such as an X-ray or ultrasound system. The purpose is to monitor both the operating status of the device and to analyse any tampering with it from a security point of view.
The platform, therefore, allows you to monitor in real time the operating status of the devices and the occurrence of any security problems due to tampering or attacks (cyber security), by notifying the control centre of the onset of these phenomena.
Sensing the future is our way to help our customers achieve impressive results.
We are a highly innovative company that exploits the most advanced technologies to monitor, predict and control complex data from different sources in real-time.
Website: https://www.sensoworks.com
LinkedIn: https://www.linkedin.com/company/sensoworks
The four mosquito-borne dengue virus serotypes (DENV1–DENV4) cause a high burden of disease throughout the tropical and sub-tropical regions of the world. Nevertheless, their precise epidemiological history in Africa - including when and where they originated and were distributed during the 20th century - remains unclear, stressing the need for One Health focused research.
The Sensoworks IIoT Platform (Intelligent Internet of Things) analyses a set of clinical data from a biochemical sensor (patch) for continuous patient monitoring.
The method is similar to Continuous Glucose Monitoring, which provides - at very short intervals -the values detected in the subcutaneous interstitial fluid, recording the variation of the parameters.
The biological element interacts with the substrate to be analysed and, thanks to the transduction system (sensor), it converts the biochemical response into an electrical signal.
Leveraging wireless radio frequency technology, the continuous monitoring system is equipped with a small and light sensor connected to a transmitter, which sends the data detected by the sensor to the patient's smartphone on regular intervals during the 24 hours.
The smartphone sends data to the Sensoworks IIoT Platform in real-time, communicating the detected parameters. These parameters will be displayed and analysed by a health worker, in order to monitor the possible evolution of the infection.
In the hypothesis of a patient who develops the disease in the days following the return from a tropical country, the Sensoworks IIoT Platform will allow the healthcare professional to identify the patient for tracking purposes (contact tracing).
The Sensoworks IIoT Platform, therefore, acts as a centralised tool for the management of the eventual evolution of the disease, for all patients under observation to whom the biochemical sensor has been applied, effectively constituting a real centre for the management and tracking of a possible epidemic.
Sensing the future is our way to help our customers achieve impressive results.
We are a highly innovative company that exploits the most advanced technologies to monitor, predict and control complex data from different sources in real-time.
Website: https://www.sensoworks.com
