Mark Crawford, Contributing Editor08.04.23
Orthopedic device manufacturers are embracing digital technologies and tools to bring real-time data and analytics to the forefront, enhancing patient outcomes and reducing healthcare costs—from initial interventional planning to surgery, tracking patient recovery, and long-term self-monitoring. In fact, digital technology is being applied in all three phases of orthopedics: pre-op, intra-op, and post-op. Several digital patient engagement platforms are already being used to engage with patients in pre-op (surgical planning phase) as well as post-op (post-surgical recovery phase).
“In addition, computer-assisted navigation and robot-assisted surgeries are also starting to enter the intra-operative phase of orthopedics,” said Neeraj Mainkar, vice president of software engineering and advanced technology at Proprio, a Seattle, Wash.-based technology company that has designed a surgical navigation system that connects pre-operative imaging and planning with intraoperative performance data. “However, we still have a long way to go before digital orthopedics becomes mainstream.”
Digital technologies such as 3D printing and web-based patient engagement platforms have now become a regular part of the orthopedic industry. Advancing steadily into the marketplace are new applications for artificial intelligence (AI)-based digital technologies, including machine learning (ML), deep learning, and natural language processing. AI is an integral part of most new applications because “it has the transformative power to truly augment ‘surgical intelligence’ and help with sophisticated diagnostics, as well as predictive analytics, with respect to post-operative patient outcomes that are truly game-changing for orthopedic surgery,” added Mainkar.
Medical device interoperability encompasses different degrees of information exchange and comprehension, as outlined by the Healthcare Information and Management Systems Society (HIMSS). The four levels—foundational, structural, semantic, and organizational—progress from basic data exchange to the integration of governance, policy, and workflow considerations.
“The complexity and functionality of interoperability extend beyond just device design and operation,” said Jennifer Samproni, chief technology officer of the Health Solutions business group at Flex, a global manufacturer serving diverse industries, including healthcare. “Interoperability in medical devices is an adaptive concept, molded by the device's functionality and design complexity. This dynamic attribute allows for effective integration and communication within the orthopedic space and many healthcare environments.”
As increasingly varied and complex digital technologies, devices, and data systems enter the healthcare field, it becomes paramount they interplay smoothly to achieve full convergence and integration. The necessity for seamless compatibility among these complex systems redefines the traditional concept of interoperability, stressing a fluid nature that varies based on the device's context and sophistication, noted Samproni. “For example,” she said, “magnetic resonance imaging [MRI] and computed tomography [CT] scanners produce intricate data sets. Interoperability for such devices does not simply mean data exchange, but also includes the correct interpretation of the data by other systems, considering the rich and nuanced information these devices generate.”
“Smart orthopedic implants integrate with the body and provide real-time data to both patients and their healthcare providers,” stated Code Technology, a Phoenix, Ariz-based healthcare software company, on its website. “They have been used to measure pressure, force, strain, stress, displacement, proximity, and temperature from inside the body. Such digital technology allows providers to monitor the health and function of the implant, enabling them to make more informed decisions about the care and treatment of their patients.”1
Companies that have developed smart orthopedic devices and implants in use today include Stryker’s Motion Sense, a wearable therapeutic monitoring device that transmits patient recovery information directly to the care team.2 This enables the personalization of a patient’s recovery by customizing physical therapy exercises based on key metrics such as pain scores, range of motion, incision images, and daily steps.
Another smart implant is Persona IQ that combines Zimmer Biomet’s knee implant Persona with Canary Medical’s implantable sensor, which received FDA approval in 2021.3 After surgery, the knee implant records and wirelessly transmits a wide range of gait data to the patient's personal base station at home. The data are then securely delivered to a cloud-based platform where surgeons can assess post-surgery recovery progress by comparing pre-operative mobility metrics with post-operative gait metrics.
Researchers at the University of Illinois have recently developed a foil coating with nanostructures for orthopedic implants that eliminates nearly all bacterial growth and can also detect when implants are starting to slip or fail by using flexible electronic sensors. Digital technologies were used to create the ultra-thin foil, including the highly sensitive, flexible electronic sensors that were attached to monitor strain. Strain changes as small as 0.1% can be precisely measured in these implants, which could indicate the implant is beginning to malfunction, allowing for earlier and hopefully simpler intervention and resolution.4
Leveraging AI, ML, and neural networks can help healthcare systems standardize data and improve health outcomes. For example, GE HealthCare’s Critical Care Suite, a collection of AI algorithms embedded on a mobile X-ray device, automatically analyzes images for critical conditions such as pneumothorax (a common but sometimes life-threatening lung disorder); when patients are identified with this condition, they are fast-tracked to a higher level of care. A recent review at University Hospital in Cleveland showed that while using Critical Care Suite, there was a 78% reduction in reporting time—from 3 hours and 22 minutes down to 44 minutes—for urgent exams based on positive detection for pneumothorax.5
According to the National Center for Biotechnology Information, there are currently four significant uses for AR in orthopedics: fracture care, adult reconstruction, oncology, and spinal surgery.6 Cameras in AR systems can take photographs of the surgical procedure in real time. Voice recognition software powered by AI allows surgeons to “call up” selected patient information as needed, certain images, or to view guidance for a specific step in the procedure. AR technology also eliminates the need for mechanical guides, making joint replacement surgery faster and more efficient.
In 2022, Surgalign’s HOLO Portal became the world’s first surgical guidance system to incorporate AI and AR.7 The device combines ML-based image guidance technology with AR, automated spine segmentation, and automated surgical planning, utilizing proprietary AI software. Intra-operative images are automatically processed by the AI system to create a patient-specific plan that is presented to the surgeon using the AR display. The AI segments and labels the anatomy to plan screw trajectories; the AR is then projected over the AI-generated plan during the surgical procedure, helping surgeons visualize the trajectories and guide their surgical instruments.
“Holo Portal is the first substantial innovation I’ve seen in the years of utilizing digital technology in my practice,” said Dr. D. Greg Anderson, MD and professor of orthopedic and neurological surgery at Thomas Jefferson University. “The system’s groundbreaking combination of AR and AI will better inform surgical decisions for my patients and ultimately deliver more accurate and efficient care in the surgical environment. It is truly transformative.”7
“In general, the FDA has a pretty good handle on regulating AI/ML devices from a public health perspective,” said Alex J. Cadotte, senior director for digital health and imaging for MCRA, a Washington, D.C.-based contract research organization and advisory firm with deep industry experience in the quality assurance, healthcare compliance, and cybersecurity markets. “However, it is important to realize that just because the FDA is not familiar with regulating a new technology, such as generative AI, that it will not cede its mission to protect public health.”
In other words, if the FDA does not understand the technology, or how it is being used in device development, the agency is unlikely to clear or approve the device. Therefore, it is often incumbent on the manufacturer to make the effort to succinctly explain (or demonstrate) how the new technology works and how it will keep the device safe and effective—"despite the newness of the technology,” said Cadotte. “If you do not, you may risk your submission failing on the first go around.”
The FDA has not recognized standards for digital health devices, especially for AI/ML device development. This is a major issue because, rather than everyone being harmonized with the same language and expectations for device performance, there is plenty of variability in what device manufacturers submit to the agency.
Because of its huge potential for disruption, AI is the hardest technology for the FDA to deal with. This is especially true if a manufacturer is trying to use “live” ML applications that engage in “active learning”—learning by analyzing data that the application is also making predictions about. “The FDA and other regulatory bodies understandably have taken a very conservative view on matters of active learning,” said Mainkar. “For this reason, most AI applications being developed for orthopedics and other critical healthcare-related applications are not active learning technologies and, in that sense, this keeps us a step behind the very latest trends in AI.”
In addition, quality control of training data is an equally important challenge that companies in digital orthopedics must prioritize. If the AI application uses labeled surgical image/video data, then having a robust validation process for that labeled data is extremely vital to ensure a smooth and accurate outcome.
Perhaps the biggest impact of device interoperability is on cybersecurity. A tension exists between the guidance for Medical Device Data Systems (MDDS) and the FDA’s focus on cybersecurity guidance over the last five years. MDDS is, by definition, about data being exchanged between interoperable devices and MDDS was down-classified by the FDA in 2011 by changing the categorization from high-risk Class III to low-risk Class I. “This was well before cybersecurity came to the forefront of FDA’s attention through recent draft guidance, which requires manufacturers to account for interoperability of MDDS device functions in their cybersecurity threat models,” said Cadotte.
Cadotte believes it is a challenge for FDA to explain how MDDS can be excluded from pre-market submissions and, at the same time, ask for MDDS functions to be included in pre-market submissions to account for cybersecurity risks as part of system architecture and threat modeling. “It can be confusing for both manufacturers and FDA review staff,” he said. “It will take time for the agency and industry to find the sweet spot on this balance.”
1. Zimmer Biomet’s ZBEdge
This ecosystem brings objective measures to the operating room. The suite of digital and robotic technologies collects and connects objective data throughout the entire episode of care. The “mymobility” digital care management platform delivers support and guidance to patients and captures continuous data and patient-reported feedback to facilitate care, outcomes, and satisfaction about patients’ surgical preparation and recovery. “We launched the WalkAI feature, the orthopedic industry’s first AI-based predictive analytics model,” said Liane Teplitsky, president of global robotics and technology and data solutions for Zimmer Biomet, a Warsaw, Ind.-based global medical technology provider of orthopedic devices, including a suite of integrated digital and robotic technologies. “Using a proprietary AI algorithm to analyze a patient’s mobility data, WalkAI helps surgeons identify patients whose recovery from hip or knee surgery may not be on track and allows them to intervene to mitigate or minimize poor outcomes. We have now also added WalkAI Patient Progress to help patients track and focus on their recovery through the smartphone app so we can see how their predicted gait speed recovery at 90 days after surgery compares with their peers.”
2. Proprio Paradigm System
Mainkar and his team are developing a navigation platform that synthesizes AI, computer vision, and machine learning to reduce inefficiencies and pain points in spinal surgery. Mainkar’s work involves training algorithms to do a host of tasks, from image and light-field data applications to vertebral segmentation. “I think the best example of digital breakthroughs in the orthopedic space is what Proprio is currently doing—combining sophisticated AI applications with real-time surgical guidance using 3D light-field technology,” said Mainkar.
Using advancements in light field, computer vision, and other emerging, intelligent technologies, Proprio’s Paradigm system synthesizes views from multiple inputs for navigating anatomy and surgical environments in 3D. Light field imaging technology captures high-definition intraoperative images to fuse with preoperative scans, which allows the surgeon to see more, do more, and achieve more, without interruption.
The Paradigm system takes the surgeons from the era of using the equivalent of a “static” MapQuest-type technology, where they had to pre-plan their surgical map-route with no real-time intra-op update, to a “Google Map” technology, where surgeons get real-time image-guided feedback about where they are during the intraoperative phase. “In applications like spinal alignment, this is a fundamental breakthrough in improving surgical outcomes,” said Mainkar.
3. Early Detection of Epileptic Seizures
Researchers at Know Biological and Sandia National Laboratories have developed a sensor that can reproduce the seizure-sniffing abilities of trained dogs that they hope can work like a wristwatch in the future.8 This advanced warning can give people with epilepsy time to take medication that can halt more seizures, or at least find a safe and private space. Digital technologies/components include a preconcentrator, sensors, and a miniature ion mobility spectrometer that analyzes gas types exiting from micro-channels in the device. The device is sensitive enough to detect even tiny amounts of gasses, such as those released by epilepsy patients during seizures.
Currently, the entire sensor is about the size of a hardcover novel—the next step is trying to reduce that size by half, and add more functionality. This will require integrated digital technologies to achieve, noted Sandia National Laboratories biomedical engineer Philip Miller.
“In a deployed wearable format, we intend for the system to autonomously collect, analyze, and alert without needing user intervention,” said Miller. “This will require cloud connectivity and algorithms to keep track of the near-constant generation of data. Miniaturizing the system is primarily an engineering challenge. Digital technologies will help us reduce the processing burden of the device and thus potentially reduce size, weight, and power requirements of the wearable module.”
“Studies suggest that 3D digital templating and operative planning achieve at least 90% accuracy, whereas traditional manual planning achieves only 57% accuracy,” said Derek Shanahan, vice president of marketing for Exer, a Denver, Colo.-based developer of AI-driven healthcare software. “Now, we are starting to see AI used to enhance orthopedic templating. Programs such as PeekMed or OrthoPlan 2.0 use AI to automate templating. These tools use AI to perform bone segmentation and landmark detection to suggest the most suitable template and its optimal position.”9
Software is also a key driver for bringing device interoperability and AI closer together. Applications include automating modeling subtasks, strengthening standardization with digitization and redefined workflows, and improving quality and speed of design and manufacturing to meet a patient’s individual needs. AI can also streamline the compliance process, aligning medical devices with necessary regulations and enabling seamless interconnectivity.
Digital innovation and convergence will continue to pick up speed, pushed along by ongoing advances in AI. Speed to market has always been a huge factor for medical device companies and AI will get their products to market that much faster. “All implants will be smart. Everything implanted in the human body will eventually have sensors, computing and communication capabilities. The information will include basic implant ID as well as real-time data about biologic processes surrounding the implant and any catastrophic failure,” stated Robin Young, president of Robin Health Care, a Berkeley, Calif.-based developer of digital documentation software for orthopedic physicians and administrators. “AI will be embedded in payor systems, hospital software, diagnostic and image equipment. As AI becomes more universal, new knowledge will be created at an explosive rate.”10
“Navigating such a complex landscape of digital convergence and device interoperability is a nuanced undertaking,” concluded Samproni. “It does present an opportunity, however, to redefine the medical device industry's contours, driving us to deliver innovative products that improve patient care.”
References
Mark Crawford is a full-time freelance business and marketing/communications writer based in Corrales, N.M. His clients range from startups to global manufacturing leaders. He has written for MPO and ODT magazines for more than 15 years and is the author of five books.
“In addition, computer-assisted navigation and robot-assisted surgeries are also starting to enter the intra-operative phase of orthopedics,” said Neeraj Mainkar, vice president of software engineering and advanced technology at Proprio, a Seattle, Wash.-based technology company that has designed a surgical navigation system that connects pre-operative imaging and planning with intraoperative performance data. “However, we still have a long way to go before digital orthopedics becomes mainstream.”
Digital technologies such as 3D printing and web-based patient engagement platforms have now become a regular part of the orthopedic industry. Advancing steadily into the marketplace are new applications for artificial intelligence (AI)-based digital technologies, including machine learning (ML), deep learning, and natural language processing. AI is an integral part of most new applications because “it has the transformative power to truly augment ‘surgical intelligence’ and help with sophisticated diagnostics, as well as predictive analytics, with respect to post-operative patient outcomes that are truly game-changing for orthopedic surgery,” added Mainkar.
Interoperability and Data Sharing
As impressive as these digital advances are, “convergence” cannot happen without effective device interoperability. Seamless data sharing is a core requirement for the full adoption and integration of digital technologies in the orthopedic space. Lack of such interoperability between different devices (including legacy) and applications is holding back the full convergence of digital technologies—not just in orthopedics, but for all surgical and healthcare applications. This increases the burden of manual documentation on medical staff and slows down the entire process of healthcare.Medical device interoperability encompasses different degrees of information exchange and comprehension, as outlined by the Healthcare Information and Management Systems Society (HIMSS). The four levels—foundational, structural, semantic, and organizational—progress from basic data exchange to the integration of governance, policy, and workflow considerations.
“The complexity and functionality of interoperability extend beyond just device design and operation,” said Jennifer Samproni, chief technology officer of the Health Solutions business group at Flex, a global manufacturer serving diverse industries, including healthcare. “Interoperability in medical devices is an adaptive concept, molded by the device's functionality and design complexity. This dynamic attribute allows for effective integration and communication within the orthopedic space and many healthcare environments.”
As increasingly varied and complex digital technologies, devices, and data systems enter the healthcare field, it becomes paramount they interplay smoothly to achieve full convergence and integration. The necessity for seamless compatibility among these complex systems redefines the traditional concept of interoperability, stressing a fluid nature that varies based on the device's context and sophistication, noted Samproni. “For example,” she said, “magnetic resonance imaging [MRI] and computed tomography [CT] scanners produce intricate data sets. Interoperability for such devices does not simply mean data exchange, but also includes the correct interpretation of the data by other systems, considering the rich and nuanced information these devices generate.”
Smart Implants
Smart implants and other devices are a rapidly growing orthopedic sector that continues to evolve quickly as designers and engineers become more aware of the seemingly unlimited applications for digital technologies that interact with each other and transmit data in real time. Smart implants are equipped with components such as sensors, microprocessors, and other electronics that provide real-time monitoring and effective personalized care, based on quality data.“Smart orthopedic implants integrate with the body and provide real-time data to both patients and their healthcare providers,” stated Code Technology, a Phoenix, Ariz-based healthcare software company, on its website. “They have been used to measure pressure, force, strain, stress, displacement, proximity, and temperature from inside the body. Such digital technology allows providers to monitor the health and function of the implant, enabling them to make more informed decisions about the care and treatment of their patients.”1
Companies that have developed smart orthopedic devices and implants in use today include Stryker’s Motion Sense, a wearable therapeutic monitoring device that transmits patient recovery information directly to the care team.2 This enables the personalization of a patient’s recovery by customizing physical therapy exercises based on key metrics such as pain scores, range of motion, incision images, and daily steps.
Another smart implant is Persona IQ that combines Zimmer Biomet’s knee implant Persona with Canary Medical’s implantable sensor, which received FDA approval in 2021.3 After surgery, the knee implant records and wirelessly transmits a wide range of gait data to the patient's personal base station at home. The data are then securely delivered to a cloud-based platform where surgeons can assess post-surgery recovery progress by comparing pre-operative mobility metrics with post-operative gait metrics.
Researchers at the University of Illinois have recently developed a foil coating with nanostructures for orthopedic implants that eliminates nearly all bacterial growth and can also detect when implants are starting to slip or fail by using flexible electronic sensors. Digital technologies were used to create the ultra-thin foil, including the highly sensitive, flexible electronic sensors that were attached to monitor strain. Strain changes as small as 0.1% can be precisely measured in these implants, which could indicate the implant is beginning to malfunction, allowing for earlier and hopefully simpler intervention and resolution.4
AI Makes It Happen
The emergence of AI promises to streamline data integration across diverse devices, enhancing the decision-making ability of healthcare providers. AI converges with other digital technologies to facilitate more efficient and effective data integration across multiple devices. In fact, Samproni indicated there are companies today that already use AI to evaluate orthopedic conditions and treatment options from scans. “By integrating data from various diagnostic devices and digitized patient records, AI can accurately and quickly identify patterns, enhancing diagnostic efficiency and effectiveness,” she said.Leveraging AI, ML, and neural networks can help healthcare systems standardize data and improve health outcomes. For example, GE HealthCare’s Critical Care Suite, a collection of AI algorithms embedded on a mobile X-ray device, automatically analyzes images for critical conditions such as pneumothorax (a common but sometimes life-threatening lung disorder); when patients are identified with this condition, they are fast-tracked to a higher level of care. A recent review at University Hospital in Cleveland showed that while using Critical Care Suite, there was a 78% reduction in reporting time—from 3 hours and 22 minutes down to 44 minutes—for urgent exams based on positive detection for pneumothorax.5
Augmented Reality
Augmented reality (AR) combines advanced optics, sensors, wireless transmission, and cloud-based processes to superimpose digital data and images on a real surface, such as smart glasses or a smart visor. When used in a surgical setting, AR can improve surgical accuracy, decrease operation times, reduce radiation exposure, and improved patient outcomes.According to the National Center for Biotechnology Information, there are currently four significant uses for AR in orthopedics: fracture care, adult reconstruction, oncology, and spinal surgery.6 Cameras in AR systems can take photographs of the surgical procedure in real time. Voice recognition software powered by AI allows surgeons to “call up” selected patient information as needed, certain images, or to view guidance for a specific step in the procedure. AR technology also eliminates the need for mechanical guides, making joint replacement surgery faster and more efficient.
In 2022, Surgalign’s HOLO Portal became the world’s first surgical guidance system to incorporate AI and AR.7 The device combines ML-based image guidance technology with AR, automated spine segmentation, and automated surgical planning, utilizing proprietary AI software. Intra-operative images are automatically processed by the AI system to create a patient-specific plan that is presented to the surgeon using the AR display. The AI segments and labels the anatomy to plan screw trajectories; the AR is then projected over the AI-generated plan during the surgical procedure, helping surgeons visualize the trajectories and guide their surgical instruments.
“Holo Portal is the first substantial innovation I’ve seen in the years of utilizing digital technology in my practice,” said Dr. D. Greg Anderson, MD and professor of orthopedic and neurological surgery at Thomas Jefferson University. “The system’s groundbreaking combination of AR and AI will better inform surgical decisions for my patients and ultimately deliver more accurate and efficient care in the surgical environment. It is truly transformative.”7
Regulatory Challenges
For the FDA, it is a bit of struggle trying to keep up with the pace of digital technology, especially AI. The FDA does not want to stand in the way of progress, but it also needs to understand how the technology works and what risks are present, which causes delays.“In general, the FDA has a pretty good handle on regulating AI/ML devices from a public health perspective,” said Alex J. Cadotte, senior director for digital health and imaging for MCRA, a Washington, D.C.-based contract research organization and advisory firm with deep industry experience in the quality assurance, healthcare compliance, and cybersecurity markets. “However, it is important to realize that just because the FDA is not familiar with regulating a new technology, such as generative AI, that it will not cede its mission to protect public health.”
In other words, if the FDA does not understand the technology, or how it is being used in device development, the agency is unlikely to clear or approve the device. Therefore, it is often incumbent on the manufacturer to make the effort to succinctly explain (or demonstrate) how the new technology works and how it will keep the device safe and effective—"despite the newness of the technology,” said Cadotte. “If you do not, you may risk your submission failing on the first go around.”
The FDA has not recognized standards for digital health devices, especially for AI/ML device development. This is a major issue because, rather than everyone being harmonized with the same language and expectations for device performance, there is plenty of variability in what device manufacturers submit to the agency.
Because of its huge potential for disruption, AI is the hardest technology for the FDA to deal with. This is especially true if a manufacturer is trying to use “live” ML applications that engage in “active learning”—learning by analyzing data that the application is also making predictions about. “The FDA and other regulatory bodies understandably have taken a very conservative view on matters of active learning,” said Mainkar. “For this reason, most AI applications being developed for orthopedics and other critical healthcare-related applications are not active learning technologies and, in that sense, this keeps us a step behind the very latest trends in AI.”
In addition, quality control of training data is an equally important challenge that companies in digital orthopedics must prioritize. If the AI application uses labeled surgical image/video data, then having a robust validation process for that labeled data is extremely vital to ensure a smooth and accurate outcome.
Perhaps the biggest impact of device interoperability is on cybersecurity. A tension exists between the guidance for Medical Device Data Systems (MDDS) and the FDA’s focus on cybersecurity guidance over the last five years. MDDS is, by definition, about data being exchanged between interoperable devices and MDDS was down-classified by the FDA in 2011 by changing the categorization from high-risk Class III to low-risk Class I. “This was well before cybersecurity came to the forefront of FDA’s attention through recent draft guidance, which requires manufacturers to account for interoperability of MDDS device functions in their cybersecurity threat models,” said Cadotte.
Cadotte believes it is a challenge for FDA to explain how MDDS can be excluded from pre-market submissions and, at the same time, ask for MDDS functions to be included in pre-market submissions to account for cybersecurity risks as part of system architecture and threat modeling. “It can be confusing for both manufacturers and FDA review staff,” he said. “It will take time for the agency and industry to find the sweet spot on this balance.”
More Successes
Nothing speaks better to digital convergence in the orthopedic device field than the following three recent successes:1. Zimmer Biomet’s ZBEdge
This ecosystem brings objective measures to the operating room. The suite of digital and robotic technologies collects and connects objective data throughout the entire episode of care. The “mymobility” digital care management platform delivers support and guidance to patients and captures continuous data and patient-reported feedback to facilitate care, outcomes, and satisfaction about patients’ surgical preparation and recovery. “We launched the WalkAI feature, the orthopedic industry’s first AI-based predictive analytics model,” said Liane Teplitsky, president of global robotics and technology and data solutions for Zimmer Biomet, a Warsaw, Ind.-based global medical technology provider of orthopedic devices, including a suite of integrated digital and robotic technologies. “Using a proprietary AI algorithm to analyze a patient’s mobility data, WalkAI helps surgeons identify patients whose recovery from hip or knee surgery may not be on track and allows them to intervene to mitigate or minimize poor outcomes. We have now also added WalkAI Patient Progress to help patients track and focus on their recovery through the smartphone app so we can see how their predicted gait speed recovery at 90 days after surgery compares with their peers.”
2. Proprio Paradigm System
Mainkar and his team are developing a navigation platform that synthesizes AI, computer vision, and machine learning to reduce inefficiencies and pain points in spinal surgery. Mainkar’s work involves training algorithms to do a host of tasks, from image and light-field data applications to vertebral segmentation. “I think the best example of digital breakthroughs in the orthopedic space is what Proprio is currently doing—combining sophisticated AI applications with real-time surgical guidance using 3D light-field technology,” said Mainkar.
Using advancements in light field, computer vision, and other emerging, intelligent technologies, Proprio’s Paradigm system synthesizes views from multiple inputs for navigating anatomy and surgical environments in 3D. Light field imaging technology captures high-definition intraoperative images to fuse with preoperative scans, which allows the surgeon to see more, do more, and achieve more, without interruption.
The Paradigm system takes the surgeons from the era of using the equivalent of a “static” MapQuest-type technology, where they had to pre-plan their surgical map-route with no real-time intra-op update, to a “Google Map” technology, where surgeons get real-time image-guided feedback about where they are during the intraoperative phase. “In applications like spinal alignment, this is a fundamental breakthrough in improving surgical outcomes,” said Mainkar.
3. Early Detection of Epileptic Seizures
Researchers at Know Biological and Sandia National Laboratories have developed a sensor that can reproduce the seizure-sniffing abilities of trained dogs that they hope can work like a wristwatch in the future.8 This advanced warning can give people with epilepsy time to take medication that can halt more seizures, or at least find a safe and private space. Digital technologies/components include a preconcentrator, sensors, and a miniature ion mobility spectrometer that analyzes gas types exiting from micro-channels in the device. The device is sensitive enough to detect even tiny amounts of gasses, such as those released by epilepsy patients during seizures.
Currently, the entire sensor is about the size of a hardcover novel—the next step is trying to reduce that size by half, and add more functionality. This will require integrated digital technologies to achieve, noted Sandia National Laboratories biomedical engineer Philip Miller.
“In a deployed wearable format, we intend for the system to autonomously collect, analyze, and alert without needing user intervention,” said Miller. “This will require cloud connectivity and algorithms to keep track of the near-constant generation of data. Miniaturizing the system is primarily an engineering challenge. Digital technologies will help us reduce the processing burden of the device and thus potentially reduce size, weight, and power requirements of the wearable module.”
Moving Forward
Software, of course, is absolutely vital for the convergence and integration of digital technologies. Computer programming will advance rapidly, in part because computational power continues to grow at exponential rates. Digital orthopedic templating, for example, which uses radiographs to create digital representations of an implant, has already been widely adopted, as it allows for more reliable and accurate sizing, positioning, and alignment of an orthopedic implant.“Studies suggest that 3D digital templating and operative planning achieve at least 90% accuracy, whereas traditional manual planning achieves only 57% accuracy,” said Derek Shanahan, vice president of marketing for Exer, a Denver, Colo.-based developer of AI-driven healthcare software. “Now, we are starting to see AI used to enhance orthopedic templating. Programs such as PeekMed or OrthoPlan 2.0 use AI to automate templating. These tools use AI to perform bone segmentation and landmark detection to suggest the most suitable template and its optimal position.”9
Software is also a key driver for bringing device interoperability and AI closer together. Applications include automating modeling subtasks, strengthening standardization with digitization and redefined workflows, and improving quality and speed of design and manufacturing to meet a patient’s individual needs. AI can also streamline the compliance process, aligning medical devices with necessary regulations and enabling seamless interconnectivity.
Digital innovation and convergence will continue to pick up speed, pushed along by ongoing advances in AI. Speed to market has always been a huge factor for medical device companies and AI will get their products to market that much faster. “All implants will be smart. Everything implanted in the human body will eventually have sensors, computing and communication capabilities. The information will include basic implant ID as well as real-time data about biologic processes surrounding the implant and any catastrophic failure,” stated Robin Young, president of Robin Health Care, a Berkeley, Calif.-based developer of digital documentation software for orthopedic physicians and administrators. “AI will be embedded in payor systems, hospital software, diagnostic and image equipment. As AI becomes more universal, new knowledge will be created at an explosive rate.”10
“Navigating such a complex landscape of digital convergence and device interoperability is a nuanced undertaking,” concluded Samproni. “It does present an opportunity, however, to redefine the medical device industry's contours, driving us to deliver innovative products that improve patient care.”
References
- bit.ly/odt230751
- bit.ly/odt230752
- bit.ly/odt230753
- bit.ly/odt230754
- bit.ly/odt230755
- bit.ly/odt230756
- bit.ly/odt230757
- bit.ly/odt230758
- bit.ly/odt230759
- bit.ly/odt230760
Mark Crawford is a full-time freelance business and marketing/communications writer based in Corrales, N.M. His clients range from startups to global manufacturing leaders. He has written for MPO and ODT magazines for more than 15 years and is the author of five books.
