Patients, healthcare service providers, hospitals, medical equipment makers, pharmaceutical companies, professionals, and various stakeholders in the ecosystem all stand to benefit from a new generation of ML driven tools.
From anatomical geometric measurements, to cancer detection, to radiology, surgery, drug discovery and genomics, the possibilities are endless. In these scenarios, ML can lead to increased operational efficiencies, extremely positive outcomes and significant cost reduction.
In addition to providing new applications that will improve the delivery of care, AI is also expected to bring improvements to hospital operations and efficiencies to hospital workflows including planning, procedure times or selecting the right exam for the right patient, all of which will help to enhance care delivery and reduce treatment costs.
Regulatory support is steadily increasing and in the US, for example, the Federal Drug Administration (FDA) is approving more and more ML methods for diagnostic assistance and other applications.
In fact, the FDA in the US has created a new regulatory framework for ML based products. This framework refers to ML techniques as, “Software as a Medical Device” (SaMD) and envisions significant benefits to quality and efficiency of care.
To support this initiative, the FDA has introduced a “predetermined change control plan” in premarket submissions which would include the types of anticipated modifications and the associated methodology to be used to implement those changes in a controlled manner.
The FDA expects commitments from medical device manufacturers on transparency and real-world performance monitoring for SaMD, as well as periodic updates on changes that were implemented as part of the approved pre-specifications and the algorithm change protocol.
This framework enables the FDA and the manufacturers to monitor a product from its premarket development to post market performance and
allows the regulatory oversight to embrace the iterative improvement power of an SaMD, while assuring patient safety.
Doing AI right, in the context of healthcare, means creating rules and regulations that ensure that it is done safely and there are still plenty of gaps and a lack of clarity on both standards and roles.
In the UK, the government has brought together regulators and organisations such as NICE, CQC and the Health Research Authority to set in place clear, innovation friendly processes and regulations needed to drive the safer use of AI in healthcare forward.
Opportunities for ML in healthcare
There’s a broad spectrum of ways that ML can be used to solve critical healthcare problems. For example, at present digital pathology, radiology, dermatology, vascular diagnostics and ophthalmology all use standard image processing techniques.
Chest x-rays are the most common radiological procedure with over 2 billion scans performed worldwide every year, that’s 548,000 scans a day. Such a huge quantity of scans imposes a heavy load on radiologists and taxes the efficiency of the workflow.
With increased computing power, new storage and devices, the amount of healthcare data that is being captured inside a hospital is fast outpacing our ability to analyse it and here are growing concerns that only a fraction of this data is being used to improve the quality and efficiency of care.
Often ML, Deep Neural Network (DNN) and Convolutional Neural Networks (CNN) methods are able to outperform radiologists in speed and accuracy, but the expertise of a radiologist is still of paramount importance.
However, under stressful conditions during a fast decision-making process, human error rate could be as high as 30%. Aiding the decision-making process with ML methods can improve the quality of result, providing the radiologists and other specialists an additional tool.
Validation of ML are today coming from multiple and very reliable sources. In one study, conducted by the Stanford ML Group, a 121-layer CNN was trained to detect pneumonia better than four radiologists.
Similarly, in multiple other studies by the National Institute of Health and other organisations, trials around early detection of cancerous pulmonary nodules for lung cancer detection using a DNN model achieved better accuracy than multiple radiologists’ diagnosis.
Though adoption in digital pathology is slower, multiple algorithm-based detections applied in a study of breast cancer compared well, and sometimes fared better than the prognosis from several pathologists. Similarly, RNN/LSTM based approaches on genome annotation predicted better results on whether single nucleotide variants are potentially pathogenic.
Many procedures within radiology, pathology, dermatology, vascular diagnostic and ophthalmology could be on large image sizes, sometimes 5 Megapixels or larger, requiring complex image processing. Also, the ML workflow can be computing and memory intensive. The predominant computation is linear algebra and demands many computations and a multitude of parameters.
This results in billions of multiply-accumulate (MAC) operations, hundreds of Megabytes of parameter data and requires a multitude of operators and a highly-distributed memory subsystem.
So, performing accurate image inferences efficiently for tissue detection or classification using traditional computational methods on PCs and GPUs are inefficient, and healthcare companies are looking for alternate techniques to address this problem.
Improved efficiency with ACAP devices
In response, Xilinx has developed a heterogenous and a highly distributed architecture that looks to solve this problem for healthcare companies. The Versal Adaptive Compute Acceleration Platform (ACAP) family of System-on-Chips (SoCs) with its adaptable Field Programmable Gate Arrays (FPGAs), integrated digital signal processors (DSPs), integrated accelerators for deep learning, SIMD VLIW engines with a highly distributed local memory architecture and multi-processor systems are known for their ability to perform massively parallel signal processing of high-speed data in close to real-time.
In addition, the Versal ACAP has multi-terabit-per-second Network on Chip (NoC) interconnect capability and an advanced AI Engine containing hundreds of tightly integrated VLIW SIMD processors. This means computing capacity can be moved beyond 100 Tera operations per second (TOPS).
These device capabilities are able to dramatically improve the efficiency of how complex healthcare ML algorithms are solved and help to significantly accelerate healthcare applications at the edge, all with less resources, cost and power. With Versal ACAP devices, support for recurrent networks could be inherent due to the simple nature of the architecture and its supporting libraries.
Xilinx has also put in place an innovative ecosystem for algorithm and application developers.
Unified software platforms, such as Vitis for application development and Vitis AI for optimising and deploying accelerated ML inference, mean developers can use advanced devices – such as ACAPs - in their projects.
Over the past decade, the growth in computational power has resulted in a massive increase in the amount and granularity of stored digital medical and healthcare data. The ability of AI to analyse large volumes of this data and create meaningful – and actionable – insights at speed, will have profound effects on how healthcare is delivered and received.
Healthcare and medical device workflows are undergoing major changes and, in the future, medical workflows will be ‘Big Data’ enterprises with significantly higher requirements for computational needs, data privacy, security, patient safety and accuracy.
Distributed, non-linear, parallel and heterogeneous computing platforms will play a key role when it comes to solving and managing this complexity.
Devices like Versal and the Vitis software platform will have an important role to play in delivering the optimised AI architectures of the future.
Author details: Subh Bhattacharya is Lead, Healthcare, Medical Devices & Sciences at Xilinx