The golden age of implementation

/ MIT Collaborative, Research Blog / By

[vc_row type=”in_container” full_screen_row_position=”middle” scene_position=”center” text_color=”dark” text_align=”left” top_padding=”30″ overlay_strength=”0.3″ shape_divider_position=”bottom”][vc_column column_padding=”no-extra-padding” column_padding_position=”all” background_color_opacity=”1″ background_hover_color_opacity=”1″ column_shadow=”none” column_border_radius=”none” width=”1/1″ tablet_text_alignment=”default” phone_text_alignment=”default” column_border_width=”none” column_border_style=”solid”][vc_column_text][nectar_dropcap color=”#3452ff”]A [/nectar_dropcap]t the last Systems Symposium organized at Johns Hopkins University, Guru Madhavan from the National Academies spoke about how engineers can influence policy and economics. In particular, he told the story about Penicillin, which I think has important lessons for us today.[/vc_column_text][/vc_column][/vc_row][vc_row type=”in_container” full_screen_row_position=”middle” scene_position=”center” text_color=”dark” text_align=”left” top_padding=”30″ overlay_strength=”0.3″ shape_divider_position=”bottom”][vc_column centered_text=”true” column_padding=”no-extra-padding” column_padding_position=”all” background_color_opacity=”1″ background_hover_color_opacity=”1″ column_shadow=”none” column_border_radius=”none” width=”1/1″ tablet_text_alignment=”default” phone_text_alignment=”default” column_border_width=”none” column_border_style=”solid”][image_with_animation image_url=”7076″ alignment=”” animation=”Fade In” img_link_large=”yes” border_radius=”none” box_shadow=”none” max_width=”100%”][vc_column_text]Alexander Fleming (Source: Wikimedia)[/vc_column_text][/vc_column][/vc_row][vc_row type=”in_container” full_screen_row_position=”middle” scene_position=”center” text_color=”dark” text_align=”left” top_padding=”30″ overlay_strength=”0.3″ shape_divider_position=”bottom”][vc_column column_padding=”no-extra-padding” column_padding_position=”all” background_color_opacity=”1″ background_hover_color_opacity=”1″ column_shadow=”none” column_border_radius=”none” width=”1/1″ tablet_text_alignment=”default” phone_text_alignment=”default” column_border_width=”none” column_border_style=”solid”][vc_column_text]Almost everyone knows that Alexander Fleming discovered Penicillin in 1928. He shared the Nobel Prize in 1945 with Howard Florey and Ernst Chain, two other researchers who looked at further development of the drug to be better used in the population. Penicillin had been discovered, the researchers had already received recognition, and for purposes of history, the story was over.

But we had no idea how to distribute or produce Penicillin for maximum impact. As late as 1942, half the Penicillin supply of the country was used up to treat one patient with sepsis. If research and development on Penicillin had stopped then, we would not have this cure available to us today.

Thankfully, Margaret Hutchinson Rousseau used deep-tank fermentation for producing Penicillin,[/vc_column_text][/vc_column][/vc_row][vc_row type=”in_container” full_screen_row_position=”middle” scene_position=”center” text_color=”dark” text_align=”left” top_padding=”30″ overlay_strength=”0.3″ shape_divider_position=”bottom”][vc_column column_padding=”no-extra-padding” column_padding_position=”all” background_color_opacity=”1″ background_hover_color_opacity=”1″ column_shadow=”none” column_border_radius=”none” width=”1/1″ tablet_text_alignment=”default” phone_text_alignment=”default” column_border_width=”none” column_border_style=”solid”][vc_column_text]

and we finally had a method that could implement the great discovery that happened over 15 years earlier. Her impact often gets a footnote, or at most a paragraph, in the history of Penicillin. But she arguably had as much of an impact as Fleming and the others on us having access to Penicillin today.

Penicillin offers us a variety of lessons. Our society has had a complicated history with antibiotics like Penicillin, with issues of misuse, resistance, and diversity. The Clinical Trials Transformation Initiative (CTTI), has a number of research areas open on how to encourage trials for antibiotics, even for specific conditions such as pneumonia.

Margaret Hutchinson Rousseau (Source: Wikimedia)

But this offers us an even more nuanced lesson. It took 17 years from the discovery of a new drug to the actual production and dissemination. This was before the days of a lengthy clinical trial system, yet it took a while before we even knew how to use the drug in a large population. It has also taken us even longer to realize what impact the spread of antibiotics will have on our population. We knew Penicillin would work wonders, and thanks to Margaret Hutchinson Rousseau it cured millions, yet it would have been great to go back to 1928 and let them all know about the issues of today.[/vc_column_text][vc_column_text]The time from 1950 to 1960 has been called the Golden Age of Antibiotic Discovery, almost half the drugs used today were discovered in that period. Time and time again, we have heard the current system promotes drug discovery. There is funding, incentives, momentum, and fame for discovering new drugs. This is truly one of the strengths of the system.

But what about implementation? For every dollar spent on research, we spend 99 cents on the discovery of new drugs. Conversations around prevention, public health, and personalized medicine all center around what impact cures have on the population and the individual. Let alone funding, there are no incentives or reasons for researchers to perform research on implementation. Again and again, we hear this in different ways. “How do we even know big data will work in the CT System.” – Curt Meinert. “We need more work on evidence generation for new drugs.” – Robert Califf. “The system needs to move towards research on implementation.” – Ayesha Kamal.

We are discovering new cures nonstop, and we are still in a crisis with the CT system. The problem isn’t discovering new drugs. The problem is we don’t know the questions we want to answer. That last sentence was roughly quoting Clay Christensen. And that is what we as a society should be focusing on. To complement the Golden Age of Discovery, we need to usher in a Golden Age of Implementation.

So how do we encourage more Margaret Hutchinson Rousseaus? She was a chemical engineer, looking to solve a medical problem. Engineering is designed to project and test the best way to bring inventions into society. Bringing in these lessons and methods from engineering offers us a great opportunity. Bayesian and machine learning, as opposed to frequentist and regression approaches. Integrated device and continuous monitoring. A culture shift towards making the patient the consumer, designing the product that will be ideal. It’s the Margaret Hutchinson Rousseau way. Dare I say it, if she was alive today, she would call herself a systems engineer.[/vc_column_text][/vc_column][/vc_row]

Assume that a decision-maker’s uncertain behavior is observed. We develop a an inverse optimization framework to impute an objective function that is robust against misspecifications of the behavior. In our model, instead of considering multiple data points, we consider an uncertainty set that encapsulates all possible realizations of the input data. We adopt this idea from robust optimization, which has been widely used for solving optimization problems with uncertain parameters. By bringing robust and inverse optimization together, we propose a robust inverse linear optimization model for uncertain input observations. We aim to find a cost vector for the underlying forward problem such that the associated error is minimized for the worst-case realization of the uncertainty in the observed solutions. That is, such a cost vector is robust in the sense that it protects against the worst misspecification of a decision-maker’s behavior. 

As an example, we consider a diet recommendation problem. Suppose we want to learn the diet patterns and preferences of a specific person and make personalized recommendations in the future. The person’s choice, even if restricted by nutritional and budgetary constraints, may be inconsistent and vary over time. Assuming the person’s behavior can be represented by an uncertainty set, it is important to find a cost vector that renders the worst-case behavior within the uncertainty set as close to optimal as possible. Note that the cost vector can have a general meaning and may be interpreted differently depending on the application (e.g., monetary cost, utility function, or preferences). Under such a cost vector, any non-worst-case diet will thus have a smaller deviation from optimality.  

We introduce a new approach that combines inverse optimization with conventional data analytics to recover the utility function of a human operator. In this approach, a set of final decisions of the operator is observed. For instance, the final treatment plans that a clinician chose for a patient or the dietary choices that a patient made to control their disease while also considering her own personal preferences. Based on these observations, we develop a new framework that uses inverse optimization to infer how the operator prioritized different trade-offs to arrive at her decision. 

We develop a new inverse optimization framework to infer the constraint parameters of a linear (forward) optimization based on multiple observations of the system. The goal is to find a feasible region for the forward problem such that all given observations become feasible and the preferred observations become optimal. We explore the theoretical properties of the model and develop computationally efficient equivalent models. We consider an array of functions to capture various desirable properties of the inferred feasible region. We apply our method to radiation therapy treatment planning—a complex optimization problem in itself—to understand the clinical guidelines that in practice are used by oncologists. These guidelines (constraints) will standardize the practice, increase planning efficiency and automation, and make high-quality personalized treatment plans for cancer patients possible.

Inverse optimization is an area of study where the purpose is to infer the unknown parameters of an optimization problem when a set of observations is available on the previous decisions made in the settings of the problem. We develop a framework to effectively and efficiently infer the cost vector of a linear optimization problem based on multiple observations on the decisions made previously. 

We then test our models in the setting of a diet problem on a data-set obtained from NHANES; The data-set is accessible via the link bellow:

https://github.com/CSSEHealthcare/Dietary-Behavior-Dataset

In radiation therapy with continuous dose delivery for Gamma Knife® Perfexion™, the dose is delivered while the radiation machine is in movement, as oppose to the conventional step-and-shoot approach which requires the unit to stop before any radiation is delivered. Continuous delivery can increase dose homogeneity and decrease treatment time. To design inverse plans, we first find a path inside the tumor volume, along which the radiation is delivered, and then find the beam durations and shapes using a mixed-integer programming optimization (MIP) model. The MIP model considers various machine-constraints as well as clinical guidelines and constraints.

Radiation therapy is frequently used in diagnosing patients with cancer. Currently, the planning of such treatments is typically done manually which is time-consuming and prone to human error. The new advancements in computational powers and treating units now allow for designing treatment plans automatically.

To design a high-quality treatment, we select the beams sizes, positions, and shapes using optimization models and approximation algorithms. The optimization models are designed to deliver an appropriate amount of dose to the tumor volume while simultaneously avoiding sensitive healthy tissues. In this project, we work on finding the best beam positions for the radiation focal points for Gamma Knife® Perfexion™, using quadratic programming and algorithms such as grassfire and sphere-packing.