Archive for the ‘Machine Learning’ Category

Qwak: Simplifying deployment and integration of machine learning … – CTech

Position: Co-founder, CEO

Founders: Alon Lev, Lior Penso, Yuval Fernbach, Ran Romano

Company description: Qwak simplifies the deployment and integration of machine learning at scale. Qwaks ML Platform empowers data science and ML engineering teams to unblock the full realization of machine learning for the business. By abstracting the complexities of model deployment, integration and optimization, Qwak brings agility and high-velocity to all ML initiatives designed to transform business, innovate and create competitive advantage.

Amount raised: $27 million

Investors: Bessemer, Leaders Fund, StageOne, Amiti

Qwak was part of the Israeli startup squad that participated in Calcalist's Mind the Tech London 2023 conference. Calcalist's "Dream Team" to London included early-stage startup companies in various fields. The startups joined the official delegation in its journey to London and took part in roundtable discussions at the event, presenting their companies to senior executives from the British and international tech industries.

Originally posted here:
Qwak: Simplifying deployment and integration of machine learning ... - CTech

Will the Raspberry Pi 5 CPU Have Built-in Machine Learning? – MUO – MakeUseOf

Raspberry Pi has been at the forefront of single-board computers (SBCs) for quite some time. However, nearly four years after the launch of Raspberry Pi 4, a new model is on the horizon.

Previous Raspberry Pi iterations generally involved faster processors, more RAM, and with the Pi 4, improved IO. However, a lot of Pis are used for AI (artificial intelligence) and ML (machine learning) purposes, leading to a lot of speculation from DIY enthusiasts about the Raspberry Pi 5's built-in machine learning capabilities.

Whether the Raspberry Pi 5 gets built-in machine learning capabilities depends a lot on what CPU the board is based around. Raspberry Pi co-founder Eben Upton teased the future of custom Pi silicon back at the tinyML Summit 2021. Since then, an imminent Raspberry Pi 5 release with massive improvements to ML is looking very likely.

Up until Raspberry Pi 4, the development team had been using ARM's Cortex processors. However, with the release of the Raspberry Pi Pico in 2021 came the RP2040, the company's first in-house SoC (system-on-chip). While it doesn't have the same power as the Raspberry Pi Zero 2 W, one of the cheapest SBCs on the market, it does provide microcontroller capabilities similar to that of an Arduino.

The Raspberry Pi 2, Pi 3, and Pi 4 have used ARM's Cortex-A7, Cortex-A53, and Cortex-A72 processors respectively. These have increased the Pi's processing capabilities over each generation, giving each progressive Pi more ML prowess. So does that mean we'll see built-in machine learning on the Raspberry Pi 5's CPU?

While there's no official word on what processor will power the Pi 5, you can be pretty sure it'll be the most ML-capable SBC in the Raspberry Pi lineup and will most likely have built-in ML support. The company's Application Specific Integrated Circuit (ASIC) team has been working since on the next iteration, which seems to be focused on lightweight accelerators for ultra-low power ML applications.

Upton's talk at tinyML Summit 2021 suggests that it might come in the form of lightweight accelerators likely running four to eight multiply-accumulates (MACs) per clock cycle. The company has also worked with ArduCam on the ArduCam Pico4ML, which brings together ML, a camera, microphones, and a screen into a Pico-sized package.

While all the details about the Raspberry Pi 5 aren't yet confirmed, if Raspberry Pi sticks to its trend of incrementally upgrading its boards, the upcoming SBC can be a rather useful board that'll check a lot of boxes for ML enthusiasts and developers looking for cheap hardware for their ML projects.

The Raspberry Pi 5 could come with built-in machine learning support, which opens up a plethora of opportunities for just about anyone to build their own ML applications with hardware that's finally able to keep up with the technology without breaking the bank.

You can already run anything from a large language model (LLM) to a Minecraft server on existing Raspberry Pis. As the SBC becomes more capable (and accessible), the possibilities of what you can do with a single credit-card-sized computer will also increase.

The rest is here:
Will the Raspberry Pi 5 CPU Have Built-in Machine Learning? - MUO - MakeUseOf

Enhancing airport operations through cloud-native technology, AI … – Airport Technology

Through its portfolio, Airport management software company AeroCloud strives to provide customers with a crystal ball, managing airports in real-time as well as predicting passenger flow and gate optimisation.

The company is developing a reputation for introducing new technology and being customer centric when offering solutions.

George Richardson, CEO of AeroCloud, has spoken with Airport Technology about the companys strategy, challenges and opportunities in the sector, and the importance of business models.

Jasleen Mann: How and when was AeroCloud founded?

George Richardson: AeroCloud was founded in Macclesfield in the North West of England in 2019 by myself and my co-founder and our CTO, Ian Forde-Smith. He has worked in the airport sector his entire career, and I was a retired racing driver, having driven competitively for 10 years.

I was excited at the prospect of bringing together Ians extensive sector and technical knowledge and my commercial mindset to create solutions that enable the airport sector to seize on the benefits of the cloud.

We built a matchbox business plan designed to compete with and displace one of five legacy operators in the airport space. Were a typical David and Goliath story and its been a tremendous journey since we first launched.

Were forever grateful that our first customers at Northwest Florida Beaches International Airport and Tampa International Airport trusted in us. And to our investors for taking the time to understand the airport technology sector and subsequently getting as excited as we are about the unforeseen potential we can bring to airports as a SaaS company.

JM: What are AeroClouds key areas of focus?

GR: AeroCloud is an intelligent management platform designed for the airport sector and the only cloud-native player in this space. The platform enables everything from faster passenger processing times to improved self-service check-in and bag drop and facilitates, increased communication between stakeholders to deal with real-time fluctuations in processes to ensure that airports work better and communicate with their customers.

We recently also launched an industry-first computer vision solution for airports that offers kerb to gate insights for the first time. AeroCloud Optic uses computer vision to intelligently, anonymously and accurately track passengers as they move through an airport. The real-time monitoring of passengers triggers alerts in response to operational bottlenecks such as extended wait times at security, which can then be immediately addressed.

The AI and machine learning algorithms also allow airport staff to identify trends and predict future scenarios to inform more accurate decision-making and long-term planning. This enables better resource management and enhanced retail opportunities for concession partners, which in turn improves the airports passenger experience.

JM: What are the challenges in this area?

GR: Airports are fast-moving complex domains, requiring smooth coordination of multiple factors in a high-pressure environment from security to passengers and airlines. A lack of synergy between these different factors can affect an airports performance in some instances, operates at only around 66% capacity.

We want to solve problems the aviation industry has struggled with for decades due to the reliance on on-site legacy technology, which isnt fit for purpose because its clunky, needs on-site maintenance, and doesnt take advantage of the latest technological innovations enabled through the cloud.

JM: What opportunities have you embraced?

GR: We have put in so much hard work to ensure that what we bring to the market is revolutionary for our sector, leveraging new technologies and practices to solve issues that have existed for decades.

We have also created an innovative ecosystem of evangelical customers, and meaning we invite them to tell our product team their problems. We then collaborate with them to find solutions and that informs the features that we create and deliver. This means that we remain customer centric across every business area, from product design and development through to customer support.

Our mission is to be the largest provider of airport operation automation software for the small to medium-sized airport market globally. In February 2023, we successfully raised $12.6million in Series A funding after we were able to demonstrate our commitment to and progress against this goal.

Indeed, when commenting on the motivation behind investing in AeroCloud, one of our new investors, Liz Christo at Stage 2 Capital, said In only a small time, AeroCloud has become the definitive operating software for small to medium-sized airports. With this new funding, we plan to deliver on our bold ambitions to expand our business, employ local people in the North West where we were founded, and continue to displace our competitors.

JM: What is the importance of democratisation of data across airports?

GR: Coordination and communication are key to operational management. Yet in many airports, most stakeholders are in the dark about the current state of play. Data is not readily accessible and many third parties might never see it beyond periodical reports.

Putting data in the hands of all relevant parties helps them understand how their services are performing and how that impacts the airport operations as a whole. That is why we offer unlimited licenses to our cloud-based platform. We dont want airports to have to choose who has access to data nor reduce its potential in supporting better operational decision-making.

JM: How does the companys strategy differ to competitors?

GR: The opportunity, we think, is a $20 billion market in which legacy players dominate; AeroCloud is the only 100% cloud-native supplier, and we are shaking up the status quo.

We can centralise all an airports operational data and flight data in about 48 hours. Our data also operates in real-time and is up to 30% more accurate than our competitors. When airports require updates or issues fixed, we deliver these via the cloud, saving resourcing and money unlike our legacy competitors who have to send a technician on site.

We also enable unlimited licenses per customer so an airports entire stakeholder base can access to the platform at no extra cost. This means AeroCloud can be used on any device wherever an airports team is based, whether thats onsite or remote that ensures the platform is more secure than legacy systems which are often run on centrally located stack servers.

JM: How do airport business models now compare to pre-pandemic models?

GR: Even before the 2020 Covid-19 pandemic, which caused escalating passenger processing times and labour shortages, these issues were difficult to manage, particularly for small and medium-sized airports that dont have the budgets and capacities that their larger peers do. As travel returns to pre-pandemic levels, airports have struggled with adjusting to heightened demand, affecting operations worldwide.

And while it is a difficult time still for many of their airports as they whole industry faces due to the debts that arose during the pandemic, it is the time when they need to invest in improving the operational efficiencies of their airports to help them boost passenger experience and revenue in the long-term.

JM: What are the implications of the 80:20 rule?

GR: The 80:20 rule requires airlines to use 80% of their take-off and landing slots or risk losing them to a competitor the following year. The rule was relaxed during the pandemic after IATA highlighted the changing schedules that many airlines were facing.

From an airport perspective, the reintroduction of this rule will necessitate a seamless journey of passengers through the terminal so that they arrive to their gate on time and support airlines to leave within their allotted timing slot.

Read more:
Enhancing airport operations through cloud-native technology, AI ... - Airport Technology

AI and Machine Learning in Wealth Management: Customized Portfolios, Predictive Analytics – Finance Magnates

Wealthmanagement is a complex and constantly evolving field, with a vast amount ofdata to analyze and complex decisions to make. With the rise of artificialintelligence (AI) and machine learning (ML), the field of wealth management hasexperienced a significant transformation in recent years.

In thisarticle, we will explore the benefits of AI and ML in wealth management,including customized portfolios and predictive analytics.

One of the mostsignificant benefits of AI and ML in wealth management is the ability to createcustomized portfolios for clients. Traditionally, wealth managers relied onmanual analysis and intuition to create investment portfolios for theirclients.

This processwas time-consuming, costly, and often resulted in portfolios that were notfully optimized for the client's unique financial situation and goals.

Keep Reading

AI and MLtechnologies can analyze vast amounts of data quickly and accurately, providingwealth managers with the insights needed to create customized investmentportfolios that meet the unique needs of each client.

Thesetechnologies can analyze factors such as risk tolerance, investment goals, andfinancial situation to create a portfolio that is tailored to the client'sspecific needs.

In addition, AIand ML can continually monitor the portfolio and adjust it as needed to ensurethat it remains aligned with the client's goals and objectives. This can helpto optimize portfolio performance and reduce the risk of losses due to marketfluctuations or other factors.

Anothersignificant benefit of AI and ML in wealth management is the ability to usepredictive analytics to make more informed investment decisions.

Predictiveanalytics involves using historical data and machine learning algorithms tomake predictions about future market trends and asset performance.

By analyzingvast amounts of data, including economic indicators, market trends, and assetperformance, AI and ML technologies can provide wealth managers with insightsand predictions that would be impossible to obtain through manual analysisalone.

Thesetechnologies can identify patterns and trends in the data that humans maynot be able to detect, providing wealth managers with a more comprehensive andaccurate view of the market.

Thisinformation can be used to make more informed investment decisions, such aswhich assets to invest in and when to buy or sell them. Predictive analyticscan also help wealth managers to identify potential risks and opportunities,allowing them to make proactive decisions to mitigate risk and capitalize onmarket opportunities.

While AI and MLtechnologies offer significant benefits for wealth management, there are alsochallenges and considerations to keep in mind. One of the primary challenges isensuring the accuracy and reliability of the data used to train the machinelearning algorithms.

If the data isbiased or incomplete, the algorithms may produce inaccurate or unreliablepredictions, leading to poor investment decisions and potential losses.

Anotherconsideration is the ethical and regulatory implications of using AI and ML inwealth management. As these technologies become increasingly sophisticated, itis essential to ensure that they are used ethically and in compliance withregulatory requirements.

This includesconsiderations such as data privacy, transparency, and accountability.

AI has theability to analyze large sets of data and provide insights that humans may notbe able to uncover. However, as with any technology, there are risks involved,and AI can backfire on wealth management in several ways.

One of the mostsignificant risks of using AI in wealth management is the potential for biasedalgorithms. AI is only as good as the data it is trained on, and if the data isbiased, the algorithms will also be biased. This can lead to unequal treatment ofclients and inaccurate investment decisions. For example, if the AI algorithmis trained on data that disproportionately represents wealthy individuals, itmay not be able to accurately predict the needs and goals of lower-incomeclients.

Moreover, AIrelies on historical data to make predictions about the future, and if thatdata is biased, the algorithm will also be biased. Biased algorithms can leadto inaccurate predictions and investment decisions, which can result infinancial losses for clients. For example, an algorithm trained on historicaldata that disproportionately represents a certain industry or demographic maynot be able to accurately predict the performance of other industries ordemographics.

While AI cananalyze vast amounts of data quickly, it cannot replace human expertise andjudgment entirely. Overreliance on technology can lead to missed opportunitiesor suboptimal investment decisions. A combination of human expertise andAI-powered analytics can lead to better investment decisions, but it isimportant to strike a balance between the two.

There is a riskthat AI can reinforce existing inequalities in wealth management. Wealthmanagement firms that use AI may be more likely to cater to wealthy clients whocan afford their services while ignoring lower-income clients. This can createa vicious cycle where wealthy clients continue to benefit from AI-poweredwealth management services, while those with less wealth are left behind.

AI and MLtechnologies are transforming the field of wealth management, providing wealthmanagers with new insights and capabilities to create customized portfolios andmake more informed investment decisions.

By analyzingvast amounts of data and using predictive analytics, these technologies canhelp wealth managers to optimize portfolio performance, reduce risk, andcapitalize on market opportunities.

However, it isessential to keep in mind the challenges and considerations associated withusing AI and ML in wealth management.

Wealth managersmust ensure the accuracy and reliability of the data used to train the machinelearning algorithms and consider the ethical and regulatory implications ofusing these technologies.

Overall, AI andML have the potential to revolutionize the field of wealth management andprovide significant benefits for both wealth managers and their clients. Asthese technologies continue to evolve, it is essential for wealth managers tostay informed and embrace them to remain competitive in a rapidly evolvingindustry.

Wealthmanagement is a complex and constantly evolving field, with a vast amount ofdata to analyze and complex decisions to make. With the rise of artificialintelligence (AI) and machine learning (ML), the field of wealth management hasexperienced a significant transformation in recent years.

In thisarticle, we will explore the benefits of AI and ML in wealth management,including customized portfolios and predictive analytics.

One of the mostsignificant benefits of AI and ML in wealth management is the ability to createcustomized portfolios for clients. Traditionally, wealth managers relied onmanual analysis and intuition to create investment portfolios for theirclients.

This processwas time-consuming, costly, and often resulted in portfolios that were notfully optimized for the client's unique financial situation and goals.

Keep Reading

AI and MLtechnologies can analyze vast amounts of data quickly and accurately, providingwealth managers with the insights needed to create customized investmentportfolios that meet the unique needs of each client.

Thesetechnologies can analyze factors such as risk tolerance, investment goals, andfinancial situation to create a portfolio that is tailored to the client'sspecific needs.

In addition, AIand ML can continually monitor the portfolio and adjust it as needed to ensurethat it remains aligned with the client's goals and objectives. This can helpto optimize portfolio performance and reduce the risk of losses due to marketfluctuations or other factors.

Anothersignificant benefit of AI and ML in wealth management is the ability to usepredictive analytics to make more informed investment decisions.

Predictiveanalytics involves using historical data and machine learning algorithms tomake predictions about future market trends and asset performance.

By analyzingvast amounts of data, including economic indicators, market trends, and assetperformance, AI and ML technologies can provide wealth managers with insightsand predictions that would be impossible to obtain through manual analysisalone.

Thesetechnologies can identify patterns and trends in the data that humans maynot be able to detect, providing wealth managers with a more comprehensive andaccurate view of the market.

Thisinformation can be used to make more informed investment decisions, such aswhich assets to invest in and when to buy or sell them. Predictive analyticscan also help wealth managers to identify potential risks and opportunities,allowing them to make proactive decisions to mitigate risk and capitalize onmarket opportunities.

While AI and MLtechnologies offer significant benefits for wealth management, there are alsochallenges and considerations to keep in mind. One of the primary challenges isensuring the accuracy and reliability of the data used to train the machinelearning algorithms.

If the data isbiased or incomplete, the algorithms may produce inaccurate or unreliablepredictions, leading to poor investment decisions and potential losses.

Anotherconsideration is the ethical and regulatory implications of using AI and ML inwealth management. As these technologies become increasingly sophisticated, itis essential to ensure that they are used ethically and in compliance withregulatory requirements.

This includesconsiderations such as data privacy, transparency, and accountability.

AI has theability to analyze large sets of data and provide insights that humans may notbe able to uncover. However, as with any technology, there are risks involved,and AI can backfire on wealth management in several ways.

One of the mostsignificant risks of using AI in wealth management is the potential for biasedalgorithms. AI is only as good as the data it is trained on, and if the data isbiased, the algorithms will also be biased. This can lead to unequal treatment ofclients and inaccurate investment decisions. For example, if the AI algorithmis trained on data that disproportionately represents wealthy individuals, itmay not be able to accurately predict the needs and goals of lower-incomeclients.

Moreover, AIrelies on historical data to make predictions about the future, and if thatdata is biased, the algorithm will also be biased. Biased algorithms can leadto inaccurate predictions and investment decisions, which can result infinancial losses for clients. For example, an algorithm trained on historicaldata that disproportionately represents a certain industry or demographic maynot be able to accurately predict the performance of other industries ordemographics.

While AI cananalyze vast amounts of data quickly, it cannot replace human expertise andjudgment entirely. Overreliance on technology can lead to missed opportunitiesor suboptimal investment decisions. A combination of human expertise andAI-powered analytics can lead to better investment decisions, but it isimportant to strike a balance between the two.

There is a riskthat AI can reinforce existing inequalities in wealth management. Wealthmanagement firms that use AI may be more likely to cater to wealthy clients whocan afford their services while ignoring lower-income clients. This can createa vicious cycle where wealthy clients continue to benefit from AI-poweredwealth management services, while those with less wealth are left behind.

AI and MLtechnologies are transforming the field of wealth management, providing wealthmanagers with new insights and capabilities to create customized portfolios andmake more informed investment decisions.

By analyzingvast amounts of data and using predictive analytics, these technologies canhelp wealth managers to optimize portfolio performance, reduce risk, andcapitalize on market opportunities.

However, it isessential to keep in mind the challenges and considerations associated withusing AI and ML in wealth management.

Wealth managersmust ensure the accuracy and reliability of the data used to train the machinelearning algorithms and consider the ethical and regulatory implications ofusing these technologies.

Overall, AI andML have the potential to revolutionize the field of wealth management andprovide significant benefits for both wealth managers and their clients. Asthese technologies continue to evolve, it is essential for wealth managers tostay informed and embrace them to remain competitive in a rapidly evolvingindustry.

Original post:
AI and Machine Learning in Wealth Management: Customized Portfolios, Predictive Analytics - Finance Magnates

Application of machine learning in predicting non-alcoholic fatty liver … – Nature.com

Aggarwal, A., Puri, K., Thangada, S., Zein, N. & Alkhouri, N. Nonalcoholic fatty liver disease in children: Recent practice guidelines, where do they take us?. Curr. Pediatr. Rev. 10(2), 151161 (2014).

Article CAS PubMed Google Scholar

Khashab, M. A., Liangpunsakul, S. & Chalasani, N. Nonalcoholic fatty liver disease as a component of the metabolic syndrome. Curr. Gastroenterol. Rep. 10(1), 7380 (2008).

Article PubMed Google Scholar

Wagenknecht, L. E. et al. Correlates and heritability of nonalcoholic fatty liver disease in a minority cohort. Obesity 17(6), 12401246 (2009).

Article CAS PubMed Google Scholar

Abdelmalek, M. F. & Diehl, A. M. Nonalcoholic fatty liver disease as a complication of insulin resistance. Med. Clin. North Am. 91(6), 11251149 (2007).

Article CAS PubMed Google Scholar

Mili, S. & timac, D. Nonalcoholic fatty liver disease/steatohepatitis: Epidemiology, pathogenesis, clinical presentation and treatment. Dig. Dis. 30(2), 158162 (2012).

Article PubMed Google Scholar

Clark, J. M., Brancati, F. L. & Diehl, A. M. The prevalence and etiology of elevated aminotransferase levels in the United States. Am. J. Gastroenterol. 98(5), 960967 (2003).

Article CAS PubMed Google Scholar

Kim, W. R., Brown, R. S. Jr., Terrault, N. A. & El-Serag, H. Burden of liver disease in the United States: Summary of a workshop. Hepatology 36(1), 227242 (2002).

Article PubMed Google Scholar

McCullough, A. J. Pathophysiology of nonalcoholic steatohepatitis. J. Clin. Gastroenterol. 40, S17S29 (2006).

CAS PubMed Google Scholar

Chalasani, N. et al. The diagnosis and management of non-alcoholic fatty liver disease: Practice Guideline by the American Association for the Study of Liver Diseases, American College of Gastroenterology, and the American Gastroenterological Association. Hepatology 55(6), 20052023 (2012).

Article PubMed Google Scholar

Ertle, J. et al. Non-alcoholic fatty liver disease progresses to hepatocellular carcinoma in the absence of apparent cirrhosis. Int. J. Cancer 128(10), 24362443 (2011).

Article CAS PubMed Google Scholar

Bellentani, S. & Marino, M. Epidemiology and natural history of non-alcoholic liver disease (NAFLD). Ann. Hepatol. 8(S1), 48 (2009).

Article Google Scholar

Patton, H. M. et al. Pediatric nonalcoholic fatty liver disease: A critical appraisal of current data and implications for future research. J. Pediatr. Gastroenterol. Nutr. 43(4), 413427 (2006).

Article PubMed Google Scholar

Shiotani, A., Motoyama, M., Matsuda, T. & Miyanishi, T. Brachial-ankle pulse wave velocity in Japanese university students. Intern. Med. 44(7), 696701 (2005).

Article PubMed Google Scholar

Razmpour, F., Abbasi, B. & Ganji, A. Evaluating the accuracy and sensitivity of anthropometric and laboratory variables in diagnosing the liver steatosis and fibrosis in adolescents with non-alcoholic fatty liver disease. J. Liver Res. Disord. Ther. 4(3), 121125 (2018).

Article Google Scholar

Bellentani, S. et al. Prevalence of and risk factors for hepatic steatosis in Northern Italy. Ann. Intern. Med. 132(2), 112119 (2000).

Article CAS PubMed Google Scholar

Omagari, K. et al. Fatty liver in non-alcoholic non-overweight Japanese adults: Incidence and clinical characteristics. J. Gastroenterol. Hepatol. 17(10), 10981105 (2002).

Article PubMed Google Scholar

Shaw, N. J., Crabtree, N. J., Kibirige, M. S. & Fordham, J. N. Ethnic and gender differences in body fat in British schoolchildren as measured by DXA. Arch. Dis. Child. 92(10), 872875 (2007).

Article PubMed PubMed Central Google Scholar

Chumlea, W. C., Siervogel, R., Roche, A. F., Webb, P. & Rogers, E. Increments across age in body composition for children 10 to 18 years of age. Hum. Biol. 55, 845852 (1983).

CAS PubMed Google Scholar

Van der Sluis, I., De Ridder, M., Boot, A., Krenning, E. & de Muinck, K.-S. Reference data for bone density and body composition measured with dual energy x ray absorptiometry in white children and young adults. Arch. Dis. Child. 87(4), 341347 (2002).

Article PubMed PubMed Central Google Scholar

Alferink, L. J. M. et al. Nonalcoholic fatty liver disease in the Rotterdam study: About muscle mass, sarcopenia, fat mass, and fat distribution. J. Bone Miner. Res. 34(7), 12541263 (2019).

Article CAS PubMed Google Scholar

He, Q. et al. Sex and race differences in fat distribution among Asian, African-American, and Caucasian prepubertal children. J. Clin. Endocrinol. Metab. 87(5), 21642170 (2002).

Article CAS PubMed Google Scholar

Pudowski, P., Matusik, H., Olszaniecka, M., Lebiedowski, M. & Lorenc, R. S. Reference values for the indicators of skeletal and muscular status of healthy Polish children. J. Clin. Densitom. 8(2), 164177 (2005).

Article PubMed Google Scholar

Yang, K. C. et al. Association of non-alcoholic fatty liver disease with metabolic syndrome independently of central obesity and insulin resistance. Sci. Rep. 6(1), 110 (2016).

Google Scholar

Balakrishnan, M. et al. Obesity and risk of nonalcoholic fatty liver disease: A comparison of bioelectrical impedance analysis and conventionally-derived anthropometric measures. Clin. Gastroenterol. Hepatol. 15(12), 19651967 (2017).

Article PubMed PubMed Central Google Scholar

Brambilla, P., Bedogni, G., Heo, M. & Pietrobelli, A. Waist circumference-to-height ratio predicts adiposity better than body mass index in children and adolescents. Int. J. Obes. 37(7), 943946 (2013).

Article CAS Google Scholar

Huang, B.-A. et al. Neck circumference, along with other anthropometric indices, has an independent and additional contribution in predicting fatty liver disease. PLoSOne 10(2), e0118071 (2015).

Article PubMed PubMed Central Google Scholar

Sookoian, S. & Pirola, C. J. Systematic review with meta-analysis: Risk factors for non-alcoholic fatty liver disease suggest a shared altered metabolic and cardiovascular profile between lean and obese patients. Aliment. Pharmacol. Ther. 46(2), 8595 (2017).

Article CAS PubMed Google Scholar

Stabe, C. et al. Neck circumference as a simple tool for identifying the metabolic syndrome and insulin resistance: Results from the Brazilian Metabolic Syndrome Study. Clin. Endocrinol. 78(6), 874881 (2013).

Article CAS Google Scholar

Subramanian, V., Johnston, R., Kaye, P. & Aithal, G. Regional anthropometric measures associated with the severity of liver injury in patients with non-alcoholic fatty liver disease. Aliment. Pharmacol. Ther. 37(4), 455463 (2013).

Article CAS PubMed Google Scholar

Borruel, S. et al. Surrogate markers of visceral adiposity in young adults: Waist circumference and body mass index are more accurate than waist hip ratio, model of adipose distribution and visceral adiposity index. PLoSOne 9(12), e114112 (2014).

Article ADS PubMed PubMed Central Google Scholar

Rankinen, T., Kim, S., Perusse, L., Despres, J. & Bouchard, C. The prediction of abdominal visceral fat level from body composition and anthropometry: ROC analysis. Int. J. Obes. 23(8), 801 (1999).

Article CAS Google Scholar

Lee, S. S. & Park, S. H. Radiologic evaluation of nonalcoholic fatty liver disease. World J. Gastroenterol. WJG 20(23), 7392 (2014).

Article PubMed Google Scholar

EskandarNejad, M. Correlation of perceived body image and physical activity in women and men according to the different levels of Body Mass Index (BMI). J. Health Promot. Manag. 2, 5940 (2013).

Google Scholar

Belghaisi-Naseri, M. et al. Plasma levels of vascular endothelial growth factor and its soluble receptor in non-alcoholic fatty liver. J. Fast. Health (2018).

Dehnavi, Z. et al. Fatty Liver Index (FLI) in predicting non-alcoholic fatty liver disease (NAFLD). Hepat. Mon. 18(2) (2018).

Birjandi, M., Ayatollahi, S. M. T., Pourahmad, S. & Safarpour, A. R. Prediction and diagnosis of non-alcoholic fatty liver disease (NAFLD) and identification of its associated factors using the classification tree method. Iran. Red Crescent Med. J. 18(11) (2016).

Islam, M., Wu, C.-C., Poly, T. N., Yang, H.-C. & Li, Y.-C.J. Applications of machine learning in fatty live disease prediction. Building Continents of Knowledge in Oceans of Data: The Future of Co-Created eHealth 166170 (IOS Press, 2018).

Google Scholar

Ma, H., Xu, C.-F., Shen, Z., Yu, C.-H. & Li, Y.-M. Application of machine learning techniques for clinical predictive modeling: A cross-sectional study on nonalcoholic fatty liver disease in China. BioMed Res. Int. 2018 (2018).

Wu, C.-C. et al. Prediction of fatty liver disease using machine learning algorithms. Comput. Methods Programs Biomed. 170, 2329 (2019).

Article PubMed Google Scholar

Gaia, S. et al. Reliability of transient elastography for the detection of fibrosis in non-alcoholic fatty liver disease and chronic viral hepatitis. J. Hepatol. 54(1), 6471 (2011).

Article PubMed Google Scholar

Sasso, M. et al. Controlled attenuation parameter (CAP): A novel VCTE guided ultrasonic attenuation measurement for the evaluation of hepatic steatosis: Preliminary study and validation in a cohort of patients with chronic liver disease from various causes. Ultrasound Med. Biol. 36(11), 18251835 (2010).

Article PubMed Google Scholar

Hsu, C. et al. Magnetic resonance vs transient elastography analysis of patients with nonalcoholic fatty liver disease: A systematic review and pooled analysis of individual participants. Clin. Gastroenterol. Hepatol. 17(4), 630637 (2019).

Article PubMed Google Scholar

Shamsi, A. et al. An uncertainty-aware transfer learning-based framework for COVID-19 diagnosis. IEEE Trans. Neural Netw. Learn. Syst. 32(4), 14081417 (2021).

Article PubMed Google Scholar

Pedregosa, F. et al. Scikit-learn: Machine learning in Python. J. Mach. Learn. Res. 12, 28252830 (2011).

MathSciNet MATH Google Scholar

Noor, N. M. et al. (eds) (Trans Tech Publ, 2015).

Google Scholar

Norazian, M. N. Comparison of linear interpolation method and mean method to replace the missing values in environmental data set (2007).

Cunningham, J. P. & Ghahramani, Z. Linear dimensionality reduction: Survey, insights, and generalizations. J. Mach. Learn. Res. 16(1), 28592900 (2015).

MathSciNet MATH Google Scholar

Onat, A. et al. Neck circumference as a measure of central obesity: Associations with metabolic syndrome and obstructive sleep apnea syndrome beyond waist circumference. Clin. Nutr. 28(1), 4651 (2009).

Article PubMed Google Scholar

Rafiei, R., Fouladi, L. & Torabi, Z. Which component of metabolic syndrome is the most important one in development of colorectal adenoma?

Albhaisi, S. Noninvasive imaging modalities in nonalcoholic fatty liver disease: Where do we stand?. EMJ 4(3), 5762 (2019).

Article Google Scholar

Ferraioli, G. & Monteiro, L. B. S. Ultrasound-based techniques for the diagnosis of liver steatosis. World J. Gastroenterol. 25(40), 6053 (2019).

Article PubMed PubMed Central Google Scholar

Khov, N., Sharma, A. & Riley, T. R. Bedside ultrasound in the diagnosis of nonalcoholic fatty liver disease. World J. Gastroenterol. WJG 20(22), 6821 (2014).

Article PubMed Google Scholar

Angulo, P. et al. Liver fibrosis, but no other histologic features, is associated with long-term outcomes of patients with nonalcoholic fatty liver disease. Gastroenterology 149(2), 389-397.e10 (2015).

Article PubMed Google Scholar

Originally posted here:
Application of machine learning in predicting non-alcoholic fatty liver ... - Nature.com