With the wide application of voice technology in the medical field, it provides a new solution for the input of medical information.
The editor counted the industry giants and startups that provide related services, and analyzed how they use voice technology to solve the pain points of electronic medical records.
Medscape surveyed 15,000 medical practitioners in the United States. Nearly two-thirds of doctors said they felt burnout (42%), depressed (15%), or both (14%). The main reasons include clinicians having to deal with various complicated medical documents (56%) and spending a lot of time entering patient information into electronic health records (24%). Voice and artificial intelligence technologies are solving this pain point.
And this will undoubtedly become a big market.
The input of electronic medical records is complicated and time-consuming: the main cause of doctor's job burnout
In the past ten years, with the popularity of Electronic Health Record (EHR) in the United States, major changes have taken place in the field of healthcare. Doctors work an average of 11 hours a day, of which 6 hours are spent processing electronic medical records, and only 1.5 hours are spent processing paper documents. However, most EHR systems are now designed as huge and complex billing systems, rather than systems that focus on patient care, including visits, hospitalizations, pharmacies, bills, reimbursements, etc., and their availability and efficiency are also affected.
The complexity and time-consuming caused by this is the primary cause of doctors’ burnout and job dissatisfaction, and it is also one of the urgent problems in the medical industry today. In September last year, a study published in Annals of Family Medicine showed that primary care physicians spend more than half of their total working hours on EHR, which means that they devote most of their energy and attention to the so-called On the "administrative" task.
And job burnout can lead to a decrease in patient satisfaction, a decrease in the quality of doctor care, and an increase in the rate of medical error, the risk of medical malpractice, and the turnover rate of physicians and employees. In addition, it is also related to physician drug abuse and physician suicide. Although there are many reasons for job burnout, such as the acquisition of medical institutions by hospitals, the increase in drug prices, the implementation of the Affordable Care Act, and the gradual transformation of payment methods to a value-based model. However, the process of recording patients' visits is tedious and time-consuming, which will affect face-to-face communication with patients and the effectiveness of clinical treatment. The explosive growth of medical data also makes it difficult for doctors to obtain and manage valuable patient information, thereby improving the health of patients.
Therefore, coping with the challenges that doctors encounter in the entire work process and optimizing the entry process of the existing electronic medical records are crucial to improving overall efficiency and quality of medical services, and reducing medical costs. Research company Technavio released a report that by 2020, global hospital expenditures will exceed 72 billion U.S. dollars, with a compound annual growth rate of 6%, and voice recognition technology is a major driving force for hospital plans.
More and more medical service providers have increased their investment in speech recognition technology. For example, Premier Health, which has five hospitals and two large medical centers, spent $1.6 million to develop speech recognition software integrated with Epic. It helps doctors reduce their workload and save about 90 minutes of time every day. Due to a more efficient workflow, this software helped Premier Health save approximately $1.3 million in medical expenses.
Voice is an increasingly popular feature, especially in the field of healthcare. DRG Digital | Manhattan Research conducted a survey of 2784 doctors, and 23% said they use voice assistants at work, such as Apple’s Siri and Amazon’s Alexa. Among them, 29% of doctors said that the voice assistant system they use is part of the EHR. These data show that as more and more developers create voice tools specifically for clinical work, voice technology will provide solutions for the transcription of medical information.
How industry giants and startups solve the pain points of electronic medical record entry
The editor combed several large companies involved in electronic medical record voice entry services-Google, Amazon, iFLYTEK, Yunzhisheng, Nuance, and emerging competitors focusing on this field-Saykara, Suki, Notable.
Layout of large companies in the field of medical voice technology
The e-commerce giant Amazon is studying how to use voice technology to provide assistance for the input and extraction of data in electronic medical records, and to achieve efficient information exchange. The Alexa application platform has lightweight medical applications from institutions such as Mayo Clinic and Libertana, which can answer medical questions, send alerts in emergencies, and help users communicate with caregivers.
The voice assistant Alexa can also be integrated into the electronic medical record to become a passive recorder. Amazon is conducting trials in hospitals across the United States, including Northwell, Mass General, and Boston's Children's Hospital. However, because Alexa does not meet HIPAA standards, the tasks completed by the software are usually limited to non-recognizable uses, such as surgeons’ checklists, patient’s disease and medication information, and hospital information. If Alexa meets the HIPAA standard, the use can be further expanded.
The medical solutions launched by Nuance, the world’s largest speech recognition technology company, have covered 72% of medical institutions in the United States, with customers in more than 30 countries around the world, and a total of more than 300 million doctor-patient communication data. Thousands of medical institutions provide services. Its product, Dragon Medical One, is dedicated to providing clinical professionals with a voice navigation file system and applications to achieve the goal of new communication with patients. The application of related technologies has greatly improved the work efficiency of doctors' diagnosis, and made the collection of patient's condition fast, flexible and accurate.
In a current AI study conducted by Google, 216,221 hospitalized cases were analyzed, involving 114,003 patients and more than 46 billion data points to create accurate and scalable predictions for various clinical scenarios. Based on this research, Google is also developing a voice recognition system for clinical records, and improves the voice transcription process of electronic health records by building an automatic voice recognition technology model.
In April 2017, iFLYTEK and Chinese Academy of Medical Sciences and Peking Union Medical College signed a comprehensive strategic cooperation framework agreement, which shows that iFlytek's smart medical-related technologies, such as the oral electronic medical record system for stomatology, are officially used in Peking Union Medical College.
Prior to the signing of this strategic cooperation agreement, the aforementioned oral electronic medical record system for dentistry had been tested and implemented. The entire system includes a medical microphone that can be clipped to the doctor’s neckline, a transmitter that can be installed in the doctor’s pocket, and a receiver that can be plugged into the doctor’s work computer. During the consultation process, the doctor only needs to dictate the patient's medical record, and the structured electronic medical record will be automatically generated on the doctor's work computer. After that, the doctor can simply modify and confirm the content of the electronic medical record, then print it and provide it to the patient, and complete the electronic file preservation.
Yunzhisheng's intelligent medical voice entry system is based on a high-performance recognition engine for the medical field. It uses voice to efficiently process a large number of text entry tasks, and interacts with the hospital's HIS and PACS systems through voice and function keys on the handheld device. Doctors can effectively avoid copy and paste operations through voice input, standardize medical record input, and increase the security of medical record input.
Currently, this system can effectively save doctors over 38% of time. Since the launch of the overall medical-oriented program, Yunzhisheng has been officially launched in more than 20 representative large-scale comprehensive tertiary hospitals across the country. These hospitals are distributed in Central China, North China, South China, and Western China, including Peking Union Medical College Hospital and Peking University. The People’s Hospital, Xijing Hospital of the Fourth Military Medical University, Shenzhen Hospital of the University of Hong Kong, etc., and about 40 hospitals are in the trial operation stage.
Different from the above-mentioned large-scale enterprises launching voice services separately, the start-ups of Saykara, Suki, and Notable are more focused on the application of voice recognition technology in electronic medical records. Among them, SayKara, founded in 2015, has a team composed of former product leaders, engineers and machine learning experts from Amazon, Microsoft, Google and Nuance. SayKara's artificial intelligence voice assistant can automatically create documents, simplify workflow, and make it easier for doctors to interact with the EHR system. The data shows that doctors who use SayKara spend 70% less time managing electronic health records, which is more conducive for them to communicate with patients and provide higher-quality medical services. At present, SayKara has cooperated with several large and medium-sized medical care systems in the United States, including OrthoIndy, a well-known plastic surgery organization as one of the early pilots.
Suki's predecessor was Robin AI. The company's artificial intelligence assistant with voice functions helped doctors reduce the burden of documents and improve the process of entering information and data. Suki has carried out 12 pilot projects in California and Georgia, involving the fields of internal medicine, ophthalmology, and plastic surgery. By using its products in three different EHR systems five days a week, preliminary results of the project showed that Suki reduced the time doctors spend on medical records by 60%. In addition, Suki also cooperates with Apple, Google, Salesforce and 23andMe to provide cutting-edge technology products for consumers, medical institutions, and large enterprises.
The product launched by Notable can automatically record doctor's consultation records and update electronic medical records. The company's solution uses natural language processing and speech recognition technology to automatically record the interaction between doctors and patients, decipher doctors' notes, and construct data structures to facilitate the entry of electronic medical records. In order to make the system run smoothly, researchers spent a lot of time recording and monitoring more than 2,000 doctor-patient interactions. Currently, Notable is developing products for Apple Watch.
Medical Voice Market: Difficulties but bright prospects
Currently, the application of voice technology in the medical field still faces three major difficulties: accuracy, security and standardization.
The first is about the accuracy of voice transcription of electronic medical records. Concerns from various parties have hindered the improvement of the overall quality of medical transcription in the past few years. For this, different companies are actively seeking solutions so that voice recognition technology can better reduce the doctor's transcription burden.
For example, Google has developed and evaluated two automatic speech recognition (ASR) methods to simplify the workflow of doctors. The first system is the CTC (connectionist temporal classificaTIon) model, which focuses on the location and sequence of the phonetic unit, and directly corresponds the phonetic with the corresponding text to achieve the classification of timing issues.
The other is the LAS (listen, attend, and spell) model, which is a multi-part neural network that converts speech into a single character of the language, and then selects subsequent items according to the previous prediction order. Each model has been trained for more than 14,000 hours of anonymous medical conversations in order to improve the accuracy of speech transcription.
The research results show that the CTC model finally achieves a word error rate of 20.1%, and most errors occur at the beginning and end of speech, and the speaker's speech time is less than one second. The LAS model finally reached 18.3% word error rate. Most of the errors occurred in the dialogue stage and had nothing to do with medical terminology.
The researchers said: "With the widespread use of electronic medical record systems, there is an increasing shortage of primary care physicians and occupational burnout rates have become higher. By optimizing the process of information extraction and analysis, ASR technology can improve the voice transcription process of electronic health records. Help doctors reduce the so-called administrative burden and provide better and more focused medical services."
For the application of voice technology in the medical field, another key challenge is how to protect patient-generated data and ensure compliance with HIPAA standards. In accordance with the US Federal and State Privacy Laws (Privacy Laws), the US Department of Health and Human Services (HHS) has formulated the Federal Health Insurance Portability and Accountability Act (HIPAA: Health Insurance Portability and Accountability Act) concerning patient safety and protecting personal privacy. . HIPAA regulations set a set of standard measures for medical staff to protect the privacy of patients. When entering information in electronic medical records, relevant regulations must be strictly followed.
Finally, the issue of standardization is involved. In 2006, the American Medical Information and Management Systems Society (Healthcare Information and Management Systems Society, HIMSS) issued the "Electronic Medical Records vs. Medical Records." Electronic Health Records: Yes, There ls a Difference" white paper, proposes the Electronic Medical Records AdopTIon Model (EMRAM, electronic medical record application model), and uses this as a basis to evaluate the level of informatization construction of medical institutions. HIMSS review revolves around the electronic medical record system, with a total of eight levels. Personalized medicine, evidence-based medicine, and evidence-based management all depend decisively on the extensive and in-depth use of modern information technology.
In addition to strict requirements on the writing, use of terminology, and coding of electronic medical records, China has also carried out the "Classification and Evaluation of the Application Level of Electronic Medical Record System" since 2010. According to relevant standards, the application level of electronic medical record system is divided into 8 categories. grade. The standards of each level include the partial requirements of the electronic medical record system and the requirements of the overall information system.
Although the application of voice technology in the field of electronic medical records still faces many obstacles, reliability, portability, and cost-effectiveness will all become key factors for medical institutions to adopt transcription tools. The medical transcription industry is considered to be one of the most promising fields in the field of medical information management because it is affected by the evolving technology.
Most medical transcription equipment consists of built-in speech recognition and memory storage systems. The increasing popularity of automatic transcription technology is expected to replace various analog devices in the near future. Factors such as the increase in the value of related medical professionals or internal transcriptionists and the increase in medical transcription outsourcing services are expected to drive market demand in the next few years.
Fume Ultra Vape,Fume Ultra Disposable Vape,Fume Ultra Disposable 2500 Puffs,Fume Ultra 2500Puffs Vape
Nanning Nuoxin Technology Co., LTD , https://www.nx-vapes.com