Stakeholder input a necessity for successful model design and development

/ 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]Even the best models are useless if they can’t be implemented to solve problems. One of the reasons why modern problems of society are so complicated is that stakeholder consensus often comes after the solution is proposed and analyzed. The economics and systems models that we work on thus need stakeholder interaction and feedback to accurately identify and quantify incentives and risks. This past week, we interacted with two other projects that involve a wide variety of stakeholders.[/vc_column_text][vc_column_text]On June 10th, our team had an extensive conversation with Pam Tenaerts, Director of the Clinical Trials Transformation Initiative (CTTI) at Duke University and an Expert Advisor to our project. For almost four hours without any break, we discussed how CTTI is aiming to improve the clinical trials process, along with current problems and proposed solutions. CTTI’s approach is to generate projects through its members who are stakeholders and/or individuals experienced with the clinical trials process. They then convene a meeting of the members who work through the problem and come up with proposed solutions. Their projects range from specific issues such as designing trials to help inform the safe use of extended-release and long-acting opioids to broader analyses such as looking at the state of clinical trials through ClinicalTrials.gov data.[/vc_column_text][vc_column_text]While CTTI’s inside-out approach helps identify problems within the clinical trials system, we start from the outside and work our way in. Being non-experts, our systems models are inspired by interactions, incentives, and risks within and outside the system. Among our broader goals, we also hope to take CTTI’s work further using our own complementary methods. For example, CTTI has created an Aggregate Analysis of the ClinicalTrials.gov database in their state of clinical trials project, and we intend to mine this data to see if we can get system-wide insights as well as use the data in our models of risk and economics.[/vc_column_text][vc_column_text]After being inspired by CTTI’s approach to integrate stakeholders, we experienced such an environment first hand at the New Models meeting organized by MIT Collaborative Initiatives. In the room were stakeholders from industry, academia, healthcare, patient advocacy, military, government, and many others to have a day-long discussion on new models for our healthcare system. The participants echoed calls for system-wide changes, tackling the risk and costs in the system, and making sure incentives were aligned in any proposed changes.

In our systems models, understanding the behavior of the stakeholders and the culture they operate in is essential to accurate analyses. These two meetings helped us learn more than any dataset would ever have.

Healthcare can indeed be described as the civil rights issue of our time. This has inspired us to go out and do our best to tackle these problems. There is a lot at stake.[/vc_column_text][/vc_column][/vc_row]

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.

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