Digital Health Criteria
Software as a Medical Device (SaMD)
Software intended for one or more medical uses that may run on different operating systems or in virtual environments. Software run on a hardware medical device is a SaMD when not part of the intended use of the hardware medical device. Software is not SaMD if it drives or controls the hardware medical device.
This can include standalone software that is intended to run on general purpose computers or mobile platforms (e.g., smartphone, tablet).
Other examples include:
- SaMD that uses the microphone of a smart device to detect interrupted breathing during sleep and sounds a tone to rouse the sleeper.
- SaMD that analyzes heart rate data intended for a clinician as an aid in diagnosis of arrhythmia.
A device or product that can identify, analyze, and use big data and large complex data sets from a variety of sources. The product extracts new and relevant information or patterns to use for medical purposes. Required for artificial intelligence devices.
Advanced Analytics may include the use of statistical modeling and analytical techniques that provide insights, predictions, and recommendations based on its analysis. In that respect, devices including Advanced Analytics may have an overlap with those including Artificial Intelligence. However, Advanced Analytics techniques typically analyze large and varied datasets that cannot normally be analyzed by humans without specialized software tools, and often discover new patterns in data.
- An imaging system conducts an analysis of a patient’s melanoma by comparing it to a repository of data from past melanoma cases (including images, diagnosis, treatment plans). The system then provides a diagnosis and generates a treatment plan for the patient.
- A software program uses data from a standard CT to create a personalized 3D model of the coronary arteries and analyzes the impact that blockages have on blood flow.
A device or product that can imitate intelligent behavior or mimics human learning and reasoning. Artificial intelligence includes machine learning, neural networks, and natural language processing. Some terms used to describe artificial intelligence include: computer-aided detection/diagnosis, statistical learning, deep learning, or smart algorithms.
One rapidly growing area of Artificial Intelligence is machine learning. Machine learning is used to design an algorithm or model without explicit programming but through the use of automated training with data (e.g., a regression function or deep learning network). Devices that include Adaptive Algorithms, i.e., algorithms that continue to learn and evolve in time, are also another area of Artificial Intelligence.
Terms or jargon used to describe artificial intelligence include computer-aided detection/diagnosis, statistical learning machines/algorithms, classifier, indicator/index/indices, support vector machine, deep learning, and smart algorithm.
- An imaging system that uses algorithms to provide diagnostic information for malignant melanoma or skin cancer in patients.
- A smart ECG device that estimates the probability of acute cardiac ischemia (ACI), a common form of heart attack.
A device or product with internet-based computing that provides computer processing resources and data on demand. The cloud is a shared pool of configurable resources (e.g., computer networks, servers, storage, applications, and services). Computing and data storage resources include: servers, operating systems, networks, software, applications, services, and storage equipment.
- SaMD being executed in the cloud.
- A mobile colposcope that stores images taken on the cloud for future retrieval and review in the doctor’s office.
- A picture archiving and communications system consists of cloud-based, web-accessible software that analyzes cardiovascular images acquired from magnetic resonance (MR) scanners.
A device or product that can prevent unauthorized access, modification, misuse, or denial of use, or the unauthorized use of information which is stored, accessed, or transferred from a medical device to an external recipient.
Examples of security functions for protection include:
- Limited access to devices or products through the authentication of users (e.g. user ID and password, smartcard, biometric).
- Use of automatic timed methods to terminate sessions within the system where appropriate for the use environment.
A device or product that can exchange and use information through an electronic interface with another medical/non-medical product, system, or device.
- An infusion pump has been designed to receive patient data from any pulse oximeter and uses this data to change infusion pump settings.
- A centralized patient monitoring system receives patient data from several devices and uses this data to command and control a ventilator to adjust pressure, volume, and flow settings that are appropriate for the patient.
Medical Device Data System (MDDS)
Hardware or software that can transfer, store, convert data formats, or display medical device data without controlling or altering the functions or parameters of any connected medical device.
- The electronic transfer or exchange of medical device data. For example, this would include software that collects output from a ventilator about a patient's CO2 level and transmits the information to a central patient data repository.
- The electronic storage and retrieval of medical device data. For example, software that stores historical blood pressure information for later review by a healthcare provider.
- The electronic conversion of medical device data from one format to another in accordance with a preset specification. For example, software that converts digital data generated by a pulse oximeter into a digital format that can be printed.
- The electronic display of medical device data. For example, software that displays a previously stored electrocardiogram for a particular patient.
Please note that MDDS does not include devices intended for active patient monitoring (i.e., any device that is intended to be relied upon in deciding to take immediate clinical action or where a timely response is required).
Mobile Medical App (MMA)
A software application that meets the definition of a medical device. The MMA transforms a mobile platform into a regulated medical device or is an accessory to a regulated medical device.
- Mobile apps that transform the mobile platform into a regulated medical device by using attachments, display screens, or sensors or by including functionalities similar to those of currently regulated medical devices
- Mobile apps that area an extension of one or more medical devices by connecting to such device(s) for the purposes of controlling the device(s) or for use in active patient monitoring or analyzing medical device data
- Mobile apps that become a regulated medical device (i.e. SaMD) by performing patient-specific analysis and providing patient-specific diagnosis, or treatment recommendations
A device or product that uses wireless communication of any form (e.g., Wi-Fi, Bluetooth, NFC) to perform at least one function.
Examples include Wi-Fi, Bluetooth, and NFC.
Novel Digital Health
A device or product that includes new, unfamiliar, or unseen digital health technology never submitted, cleared, or approved by FDA. The technology could potentially be a de Novo, have a new intended use, or have different technological characteristics. This also includes digital health technology or topic areas that have no agreed upon or established definition by industry or FDA.
Examples of novel digital health technologies include but are not limited to:
- Virtual Reality
- Medical Body Area Network (MBAN) wearable or implanted wireless devices