Search for:
Find Out If Your Job Will Be Automated

The most recent jobs report, showing a high unemployment rate in 2018, also portrays the global economy capable of generating plenty of work. But what if, in the not-too-distant future, there won’t be enough jobs to go around?

That’s what some economists believe will happen as robots and artificial intelligence increasingly becomes capable of performing human tasks. Researchers, for example, estimate that nearly half of all jobs may be at risk in the coming decades, with lower-paid occupations among the most vulnerable.

Wondering how vulnerable your job might be? Refer to the chart below to see what the researchers think is the probability of your job being automated.

Automated system identifies dense tissue, a risk factor for breast cancer, in mammograms

Deep-learning model has been used successfully on patients, may lead to more consistent screening procedures.

Researchers from MIT and Massachusetts General Hospital have developed an automated model that assesses dense breast tissue in mammograms — which is an independent risk factor for breast cancer — as reliably as expert radiologists.

This marks the first time a deep-learning model of its kind has successfully been used in a clinic on real patients, according to the researchers. With broad implementation, the researchers hope the model can help bring greater reliability to breast density assessments across the nation.

It’s estimated that more than 40 percent of U.S. women have dense breast tissue, which alone increases the risk of breast cancer. Moreover, dense tissue can mask cancers on the mammogram, making screening more difficult. As a result, 30 U.S. states mandate that women must be notified if their mammograms indicate they have dense breasts.

But breast density assessments rely on subjective human assessment. Due to many factors, results vary — sometimes dramatically — across radiologists. The MIT and MGH researchers trained a deep-learning model on tens of thousands of high-quality digital mammograms to learn to distinguish different types of breast tissue, from fatty to extremely dense, based on expert assessments. Given a new mammogram, the model can then identify a density measurement that closely aligns with expert opinion.

“Breast density is an independent risk factor that drives how we communicate with women about their cancer risk. Our motivation was to create an accurate and consistent tool, that can be shared and used across health care systems,” says Adam Yala, a PhD student in MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) and second author on a paper describing the model that was published today in Radiology.

The other co-authors are first author Constance Lehman, professor of radiology at Harvard Medical School and the director of breast imaging at the MGH; and senior author Regina Barzilay, the Delta Electronics Professor at CSAIL and the Department of Electrical Engineering and Computer Science at MIT and a member of the Koch Institute for Integrative Cancer Research at MIT.

Mapping density

The model is built on a convolutional neural network (CNN), which is also used for computer vision tasks. The researchers trained and tested their model on a dataset of more than 58,000 randomly selected mammograms from more than 39,000 women screened between 2009 and 2011. For training, they used around 41,000 mammograms and, for testing, about 8,600 mammograms.

Each mammogram in the dataset has a standard Breast Imaging Reporting and Data System (BI-RADS) breast density rating in four categories: fatty, scattered (scattered density), heterogeneous (mostly dense), and dense. In both training and testing mammograms, about 40 percent were assessed as heterogeneous and dense.

During the training process, the model is given random mammograms to analyze. It learns to map the mammogram with expert radiologist density ratings. Dense breasts, for instance, contain glandular and fibrous connective tissue, which appear as compact networks of thick white lines and solid white patches. Fatty tissue networks appear much thinner, with gray area throughout. In testing, the model observes new mammograms and predicts the most likely density category.

Matching assessments

The model was implemented at the breast imaging division at MGH. In a traditional workflow, when a mammogram is taken, it’s sent to a workstation for a radiologist to assess. The researchers’ model is installed in a separate machine that intercepts the scans before it reaches the radiologist, and assigns each mammogram a density rating. When radiologists pull up a scan at their workstations, they’ll see the model’s assigned rating, which they then accept or reject.

“It takes less than a second per image … [and it can be] easily and cheaply scaled throughout hospitals.” Yala says.

On over 10,000 mammograms at MGH from January to May of this year, the model achieved 94 percent agreement among the hospital’s radiologists in a binary test — determining whether breasts were either heterogeneous and dense, or fatty and scattered. Across all four BI-RADS categories, it matched radiologists’ assessments at 90 percent. “MGH is a top breast imaging center with high inter-radiologist agreement, and this high quality dataset enabled us to develop a strong model,” Yala says.

In general testing using the original dataset, the model matched the original human expert interpretations at 77 percent across four BI-RADS categories and, in binary tests, matched the interpretations at 87 percent.

In comparison with traditional prediction models, the researchers used a metric called a kappa score, where 1 indicates that predictions agree every time, and anything lower indicates fewer instances of agreements. Kappa scores for commercially available automatic density-assessment models score a maximum of about 0.6. In the clinical application, the researchers’ model scored 0.85 kappa score and, in testing, scored a 0.67. This means the model makes better predictions than traditional models.

In an additional experiment, the researchers tested the model’s agreement with consensus from five MGH radiologists from 500 random test mammograms. The radiologists assigned breast density to the mammograms without knowledge of the original assessment, or their peers’ or the model’s assessments. In this experiment, the model achieved a kappa score of 0.78 with the radiologist consensus.

Next, the researchers aim to scale the model into other hospitals. “Building on this translational experience, we will explore how to transition machine-learning algorithms developed at MIT into clinic benefiting millions of patients,” Barzilay says.

Without emotional intelligence, artificial intelligence isn’t so smart

Vahé Torossian – Corporate Vice President, Microsoft Corporation & President Microsoft Western Europe

I truly believe that Artificial Intelligence (AI) carries enormous potential to make the world a better place and drive transformational change in some of the most important aspects of our lives – how we live, play, interact with each other and, last but not least, work. AI is also full of surprises. When it comes to AI in the workplace, we’ve noticed that a key and common trait of companies that use AI successfully is actually — Emotional Intelligence (EQ).

Simply defined, EQ is the ability to identify and understand emotions. In a professional context, it means having the ability to handle relationships with empathy, understand what motivates people, and creating an open and collaborative environment (with technology or other tools) that empowers people to do their best work. Throughout my career, I’ve seen first-hand how EQ can take an organization to a whole new level.

We recently learned a lot about AI, and its fascinating relation with EQ, from a new study commissioned by Microsoft and conducted by Ernst & Young. Released publicly today, Artificial Intelligence in Europe seeks to help us understand the AI strategies of 277 major companies across seven business sectors and 15 countries in Europe. It examines how ready these companies are to adopt AI, how the organizations rate the impact and benefits from AI implementations, and what they perceive as the keys to success.

The research found a strong and clear correlation between the maturity of AI deployments by European companies and how those organizations rate themselves on EQ. A solid majority—80 percent—of the companies most advanced in AI considered themselves to be “strongly emotionally intelligent”. On the flip side, only 16 percent of the companies considered least mature in AI rated themselves as more than moderately competent in EQ.

The correlation between EQ and AI is not obvious. Researchers asked companies about eight organizational capabilities considered necessary to successfully harness AI. EQ was rated as the least important. Companies that are less mature in their use of AI are often focused on more immediate needs such as data management and advanced analytics, which organizations ranked as the most important capabilities considered by the study.

This makes a lot of sense to me. At Microsoft we believe the promise of AI lies in what it can do to amplify our ingenuity – the power of human PLUS machine. It may well be that you must have an emotionally intelligent business culture, open to change and to new ways of working, to successfully use AI. Think about it… since AI is relatively new for most organizations, solving business and customer challenges with AI often requires that systems be designed and built from the ground up. Doing that requires business acumen, technological savvy, and a willingness to embrace the unknown. Another interesting finding in the research was that 57 percent of the companies interviewed expect that AI will have a high impact on business areas that are entirely unknown to the company today. Business leaders who are ready to embrace and tackle the unknown are already demonstrating the kind of openness that is fundamental to EQ.

I believe that businesses can open the door for tremendous growth opportunities by fostering a culture that is emotionally intelligent and that empowers workers with AI tools. 61 percent of the 277 companies included in the research expect AI to free up employees to do what people do best: think creatively and figure out how to use technological tools to drive business success, optimize operations, and engage customers in new and exciting ways. It’s hard to imagine a business leader that doesn’t want that sort of outcome. One of our challenges is that, in Europe, we have a long way to go to get there; just four percent of respondents said that AI is contributing to company-wide processes today. So all I see is opportunity!

Technology for technology’s sake has never been the answer. We must be aligned on the ultimate impact technology can have and be working toward clear and common goals – for the benefit of our customers and our society. To impact real change and to harvest the huge potential of AI, leaders must establish that clarity and create an open and collaborative culture that supports people with intelligent technology, so they can bring their very best to work every day. Then we will truly see the power of AI, to create a better tomorrow.

Impossible To Ignore: The Importance Of IT Governance

Effective IT governance is a critical tool for CIOs to align their organizations and efforts to support business strategy and create shareholder value. Given the rapidly changing and evolving technology options that confront CIOs and business leaders, making sure the right decisions are being made about investments in IT is an essential priority.

There are many misconceptions about what constitutes a comprehensive IT governance model and how it is implemented. IT governance is more than just:

  • Having a steering committee that meets periodically to review and approve IT plans and budgets
  • Involving the business on an annual basis to assist in assigning IT priorities
  • Using financial metrics such as ROI to determine whether to invest in specific initiatives
  • Instituting best practices to ensure projects are completed on time and within budget
  • Measuring and reporting on user satisfaction of IT services

While all of the above are important yardsticks to assess the impact of IT, taken one by one they do not guarantee that IT is contributing to the type of business performance that provides a competitive advantage and achieves enterprise business goals. Most of all, they do not constitute an effective IT governance program.

How Best to Think About IT Governance

IT governance comprises a decision framework and set of processes that allow CIOs and management to articulate desired outcomes through programs that enable the organization to attain these results. The decision framework and the corresponding tools and processes to support them must be clearly communicated so that day-to-day activities and decisions are made within this context. In other words, IT governance needs to instil behaviour and awareness that is understood at all levels in the organization, not just by senior management.

Clearly, the desired outcomes that shape IT will vary between industries and organizations. For example, some enterprises may focus on product innovation and accelerated go-to-market strategies while others may strive to create operational efficiencies throughout the value chain. CIOs may also encourage management to consider new technologies such as a big data, real-time analytics initiative or social-media-based customer satisfaction programs to support business performance.

The essential success factor, regardless of the specific initiative undertaken, is the linkage to tangible, measurable top-line or bottom-line business outcomes. As tempting as the latest technology or trend might be, organizations must always calibrate their IT endeavours against this metric to ensure they are not investing financial and human capital where it will yield minimal return and offer no strategic value.

CIOs must take the lead in helping place the organization’s competitive model within the governance-making framework so that the right decisions are being made and, ultimately, institutionalized across the enterprise. From a top-down perspective this means:

  • Linking business strategy to the IT programs that will be undertaken and funded; this will be reflected and communicated within the IT planning process.
  • Aligning IT spend and investments to ensure that they reflect the appropriate strategic initiatives. This is a continual process and not just part of the annual budgeting cycle.
  • Staffing the IT organization with the necessary skills and resources to effectively execute the committed programs.
  • Implementing effective risk management processes that ensure regulatory compliance, accountability, transparency and resiliency.
  • Creating a financial scorecard that tracks approved IT investments to each desired outcome measured in delivered business benefits.

Institutionalizing IT Governance

The missing link between a well-thought-out plan endorsed by management and actualization is often the absence of tactical processes and policies both inside and outside of IT. Some critical and foundational disciplines include:

Business Integration

The emergence of enterprise architectures – solutions that support end-to-end business processes – require CIOs to advocate for far greater business involvement than was traditionally required for “siloed” applications. Prerequisites are (1) business sponsorship at an executive level to provide the sense of urgency and commitment of mindshare and resources required and (2) business process owners who oversee and control the impact of new technology throughout the organization. Without these two ingredients, any strategic project will be viewed as IT-centric with little accountability from the business and, by extension, limited commitment to the desired outcomes.

Program Management

Consistent practices in managing IT projects and delivering solutions within agreed-upon parameters is a basic building block for most organizations. However, within the broader context of an IT governance framework, program management must incorporate metrics that were used within the governance framework. This would include not only the investment analysis but also the desired outcomes that drove the decision-making process. These metrics can be incorporated within dashboards that will help management view progress, benefits and the effectiveness of their decisions. As all effective management practices, IT governance needs to be continually reviewed, assessed and refined with proper measurement and transparency. Program management is essential to bridging decisions to the execution of strategic plans.

Portfolio Management

By their very nature, IT architectures consist of numerous technical layers and components, making them difficult to relate to in terms of business activities and decision making. Portfolios are useful tools for CIOs to integrate views of IT services and solutions to senior management so that they can be associated with desired outcomes. Often this will result in moving toward architecture standardization as an added dividend that will yield long-term benefits. The move to integrated solutions across the enterprise will require a restructuring of the portfolio as an overall strategy that will reduce IT costs and deliver greater operational efficiencies. The portfolio dimension is another key criteria that must be incorporated within management scorecards to guide future technology investments.

How Best to Move Forward

Developing a comprehensive IT governance program can be a daunting task even for organizations with mature management practices. The best place to start is to become familiar with the COBIT 5 framework and principles. ISACA (Information Systems Audit and Control Association) offers many valuable tools and information that will help with education and putting into place a roadmap for the IT governance journey.

Additionally, consider utilizing an experienced practitioner that can help implement practical and proven strategies to formulate an IT governance program and roadmap.  They can also assist in engaging senior management in adopting the necessary practices that will lead to acceptance across the broader organization.

It cannot be stressed enough that IT governance is an ongoing journey that will continually evolve, not a one-time destination.  It is up to CIOs to lead the way by helping their organizations think about, evaluate and adopt the “right” IT strategies for their businesses.