Limits...
From data to optimal decision making: a data-driven, probabilistic machine learning approach to decision support for patients with sepsis.

Tsoukalas A, Albertson T, Tagkopoulos I - JMIR Med Inform (2015)

Bottom Line: Furthermore, the percentage of transitions within a trajectory that lead to a better or better/same state are significantly higher by following the policy than for non-policy cases (605 vs 344 patients, P=8.6e-25).Mortality was predicted with an AUC of 0.7 and 0.82 accuracy in the general case and similar performance was obtained for the inference of the length-of-stay (AUC of 0.69 to 0.73 with accuracies from 0.69 to 0.82).A data-driven model was able to suggest favorable actions, predict mortality and length of stay with high accuracy.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Computer Science and Genome Center, University of California, Davis, Davis, CA, United States.

ABSTRACT

Background: A tantalizing question in medical informatics is how to construct knowledge from heterogeneous datasets, and as an extension, inform clinical decisions. The emergence of large-scale data integration in electronic health records (EHR) presents tremendous opportunities. However, our ability to efficiently extract informed decision support is limited due to the complexity of the clinical states and decision process, missing data and lack of analytical tools to advice based on statistical relationships.

Objective: Development and assessment of a data-driven method that infers the probability distribution of the current state of patients with sepsis, likely trajectories, optimal actions related to antibiotic administration, prediction of mortality and length-of-stay.

Methods: We present a data-driven, probabilistic framework for clinical decision support in sepsis-related cases. We first define states, actions, observations and rewards based on clinical practice, expert knowledge and data representations in an EHR dataset of 1492 patients. We then use Partially Observable Markov Decision Process (POMDP) model to derive the optimal policy based on individual patient trajectories and we evaluate the performance of the model-derived policies in a separate test set. Policy decisions were focused on the type of antibiotic combinations to administer. Multi-class and discriminative classifiers were used to predict mortality and length of stay.

Results: Data-derived antibiotic administration policies led to a favorable patient outcome in 49% of the cases, versus 37% when the alternative policies were followed (P=1.3e-13). Sensitivity analysis on the model parameters and missing data argue for a highly robust decision support tool that withstands parameter variation and data uncertainty. When the optimal policy was followed, 387 patients (25.9%) have 90% of their transitions to better states and 503 patients (33.7%) patients had 90% of their transitions to worse states (P=4.0e-06), while in the non-policy cases, these numbers are 192 (12.9%) and 764 (51.2%) patients (P=4.6e-117), respectively. Furthermore, the percentage of transitions within a trajectory that lead to a better or better/same state are significantly higher by following the policy than for non-policy cases (605 vs 344 patients, P=8.6e-25). Mortality was predicted with an AUC of 0.7 and 0.82 accuracy in the general case and similar performance was obtained for the inference of the length-of-stay (AUC of 0.69 to 0.73 with accuracies from 0.69 to 0.82).

Conclusions: A data-driven model was able to suggest favorable actions, predict mortality and length of stay with high accuracy. This work provides a solid basis for a scalable probabilistic clinical decision support framework for sepsis treatment that can be expanded to other clinically relevant states and actions, as well as a data-driven model that can be adopted in other clinical areas with sufficient training data.

No MeSH data available.


Related in: MedlinePlus

Performance and Robustness of the POMDP Clinical Decision Support System for Sepsis. (A) 5-fold cross-validation results depict the performance of each fold with (left, blue) and without (right, brown) using the policy-proposed antibiotic combination. Each cell contains the number of transitions (4225 total transitions; 843 transitions per test fold) that lead to worse, equal or better states in each case. A state-specific percentage across all folds allows for comparison between the different policy strategies. (B) Number of patient trajectories vs. the percentage of their transitions that lead to a better state. (C) Dependency of CDSS performance on data size based on stratified reduction of the dataset. Outcome is shown for policy-proposed antibiotic combinations (left) and all other combinations (right) for different states (D) Changes in antibiotic combinations proposed in calculated optimal policy as a function of data-size reduction. Each row is a state and each column represents a drug-training set combination. The two tables depict which drug combinations were found to lead to better outcomes when in the perspective state, in the general case. States are as defined in the Methods sections, with three states denoting uncertainty due to missing data (PS: probable sepsis; BPS: bacteremia, probable sepsis; PSS: probable septic shock).
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4376114&req=5

figure6: Performance and Robustness of the POMDP Clinical Decision Support System for Sepsis. (A) 5-fold cross-validation results depict the performance of each fold with (left, blue) and without (right, brown) using the policy-proposed antibiotic combination. Each cell contains the number of transitions (4225 total transitions; 843 transitions per test fold) that lead to worse, equal or better states in each case. A state-specific percentage across all folds allows for comparison between the different policy strategies. (B) Number of patient trajectories vs. the percentage of their transitions that lead to a better state. (C) Dependency of CDSS performance on data size based on stratified reduction of the dataset. Outcome is shown for policy-proposed antibiotic combinations (left) and all other combinations (right) for different states (D) Changes in antibiotic combinations proposed in calculated optimal policy as a function of data-size reduction. Each row is a state and each column represents a drug-training set combination. The two tables depict which drug combinations were found to lead to better outcomes when in the perspective state, in the general case. States are as defined in the Methods sections, with three states denoting uncertainty due to missing data (PS: probable sepsis; BPS: bacteremia, probable sepsis; PSS: probable septic shock).

Mentions: We performed 5-fold cross validation (CV) to evaluate the generalization error of our approach, with similar results across all folds, Figure 6 (a) and Tables 2 and 3. The CDSS-derived optimal policies in each transition led in significantly more occurrences to better states than when the treatment that was followed by the physicians was not in agreement with the CDSS-derived policy (49% of transitions to better states when the CDSS-derived policy was followed vs 37% otherwise; P=1.3e-13). Interestingly, when non-CDSS policies were used, patients tend to stay in the same condition (35% in non-CDSS policies vs 25% in CDSS policies, P=5.1e-13) while the difference between CDSS-derived policies and non-CDSS policies is not statistically significant for transitions to a worse condition (28% in non-CDSS policies vs 26% in CDSS policies, P=4.2e-1). We then analyzed each patient trajectory independently, to estimate the number of transitions within a trajectory that lead to better states, with and without following the policy. Results show that when the optimal policy prediction (ie, the policy that maximizes the expected cumulated reward, as defined in the methods section) is followed, there exists a significant shift to trajectories that have more than 90% of their transitions leading to a better state, Figure 6 (b). When the policy was followed, 387 patients (25.9%) have 90% of their transitions to better states and 503 patients (33.7%) patients had 90% of their transitions to worse states (P=4.0e-06), while in the non-policy cases, these numbers are 192 (12.9%) and 764 (51.2%) patients (P=4.6e-117), respectively. Furthermore, the percentage of transitions within a trajectory that lead to a better or better/same state are significantly higher by following the policy than for non-policy cases. Indeed, 605 versus 344 patients (P=8.6e-25) have 90% of their transitions to a better state with versus without following a CDSS-derived policy, Figure 1 (a). This result was observed in all five runs of the 5-fold CV and on the full dataset, hence it holds for different data distributions.


From data to optimal decision making: a data-driven, probabilistic machine learning approach to decision support for patients with sepsis.

Tsoukalas A, Albertson T, Tagkopoulos I - JMIR Med Inform (2015)

Performance and Robustness of the POMDP Clinical Decision Support System for Sepsis. (A) 5-fold cross-validation results depict the performance of each fold with (left, blue) and without (right, brown) using the policy-proposed antibiotic combination. Each cell contains the number of transitions (4225 total transitions; 843 transitions per test fold) that lead to worse, equal or better states in each case. A state-specific percentage across all folds allows for comparison between the different policy strategies. (B) Number of patient trajectories vs. the percentage of their transitions that lead to a better state. (C) Dependency of CDSS performance on data size based on stratified reduction of the dataset. Outcome is shown for policy-proposed antibiotic combinations (left) and all other combinations (right) for different states (D) Changes in antibiotic combinations proposed in calculated optimal policy as a function of data-size reduction. Each row is a state and each column represents a drug-training set combination. The two tables depict which drug combinations were found to lead to better outcomes when in the perspective state, in the general case. States are as defined in the Methods sections, with three states denoting uncertainty due to missing data (PS: probable sepsis; BPS: bacteremia, probable sepsis; PSS: probable septic shock).
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4376114&req=5

figure6: Performance and Robustness of the POMDP Clinical Decision Support System for Sepsis. (A) 5-fold cross-validation results depict the performance of each fold with (left, blue) and without (right, brown) using the policy-proposed antibiotic combination. Each cell contains the number of transitions (4225 total transitions; 843 transitions per test fold) that lead to worse, equal or better states in each case. A state-specific percentage across all folds allows for comparison between the different policy strategies. (B) Number of patient trajectories vs. the percentage of their transitions that lead to a better state. (C) Dependency of CDSS performance on data size based on stratified reduction of the dataset. Outcome is shown for policy-proposed antibiotic combinations (left) and all other combinations (right) for different states (D) Changes in antibiotic combinations proposed in calculated optimal policy as a function of data-size reduction. Each row is a state and each column represents a drug-training set combination. The two tables depict which drug combinations were found to lead to better outcomes when in the perspective state, in the general case. States are as defined in the Methods sections, with three states denoting uncertainty due to missing data (PS: probable sepsis; BPS: bacteremia, probable sepsis; PSS: probable septic shock).
Mentions: We performed 5-fold cross validation (CV) to evaluate the generalization error of our approach, with similar results across all folds, Figure 6 (a) and Tables 2 and 3. The CDSS-derived optimal policies in each transition led in significantly more occurrences to better states than when the treatment that was followed by the physicians was not in agreement with the CDSS-derived policy (49% of transitions to better states when the CDSS-derived policy was followed vs 37% otherwise; P=1.3e-13). Interestingly, when non-CDSS policies were used, patients tend to stay in the same condition (35% in non-CDSS policies vs 25% in CDSS policies, P=5.1e-13) while the difference between CDSS-derived policies and non-CDSS policies is not statistically significant for transitions to a worse condition (28% in non-CDSS policies vs 26% in CDSS policies, P=4.2e-1). We then analyzed each patient trajectory independently, to estimate the number of transitions within a trajectory that lead to better states, with and without following the policy. Results show that when the optimal policy prediction (ie, the policy that maximizes the expected cumulated reward, as defined in the methods section) is followed, there exists a significant shift to trajectories that have more than 90% of their transitions leading to a better state, Figure 6 (b). When the policy was followed, 387 patients (25.9%) have 90% of their transitions to better states and 503 patients (33.7%) patients had 90% of their transitions to worse states (P=4.0e-06), while in the non-policy cases, these numbers are 192 (12.9%) and 764 (51.2%) patients (P=4.6e-117), respectively. Furthermore, the percentage of transitions within a trajectory that lead to a better or better/same state are significantly higher by following the policy than for non-policy cases. Indeed, 605 versus 344 patients (P=8.6e-25) have 90% of their transitions to a better state with versus without following a CDSS-derived policy, Figure 1 (a). This result was observed in all five runs of the 5-fold CV and on the full dataset, hence it holds for different data distributions.

Bottom Line: Furthermore, the percentage of transitions within a trajectory that lead to a better or better/same state are significantly higher by following the policy than for non-policy cases (605 vs 344 patients, P=8.6e-25).Mortality was predicted with an AUC of 0.7 and 0.82 accuracy in the general case and similar performance was obtained for the inference of the length-of-stay (AUC of 0.69 to 0.73 with accuracies from 0.69 to 0.82).A data-driven model was able to suggest favorable actions, predict mortality and length of stay with high accuracy.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Computer Science and Genome Center, University of California, Davis, Davis, CA, United States.

ABSTRACT

Background: A tantalizing question in medical informatics is how to construct knowledge from heterogeneous datasets, and as an extension, inform clinical decisions. The emergence of large-scale data integration in electronic health records (EHR) presents tremendous opportunities. However, our ability to efficiently extract informed decision support is limited due to the complexity of the clinical states and decision process, missing data and lack of analytical tools to advice based on statistical relationships.

Objective: Development and assessment of a data-driven method that infers the probability distribution of the current state of patients with sepsis, likely trajectories, optimal actions related to antibiotic administration, prediction of mortality and length-of-stay.

Methods: We present a data-driven, probabilistic framework for clinical decision support in sepsis-related cases. We first define states, actions, observations and rewards based on clinical practice, expert knowledge and data representations in an EHR dataset of 1492 patients. We then use Partially Observable Markov Decision Process (POMDP) model to derive the optimal policy based on individual patient trajectories and we evaluate the performance of the model-derived policies in a separate test set. Policy decisions were focused on the type of antibiotic combinations to administer. Multi-class and discriminative classifiers were used to predict mortality and length of stay.

Results: Data-derived antibiotic administration policies led to a favorable patient outcome in 49% of the cases, versus 37% when the alternative policies were followed (P=1.3e-13). Sensitivity analysis on the model parameters and missing data argue for a highly robust decision support tool that withstands parameter variation and data uncertainty. When the optimal policy was followed, 387 patients (25.9%) have 90% of their transitions to better states and 503 patients (33.7%) patients had 90% of their transitions to worse states (P=4.0e-06), while in the non-policy cases, these numbers are 192 (12.9%) and 764 (51.2%) patients (P=4.6e-117), respectively. Furthermore, the percentage of transitions within a trajectory that lead to a better or better/same state are significantly higher by following the policy than for non-policy cases (605 vs 344 patients, P=8.6e-25). Mortality was predicted with an AUC of 0.7 and 0.82 accuracy in the general case and similar performance was obtained for the inference of the length-of-stay (AUC of 0.69 to 0.73 with accuracies from 0.69 to 0.82).

Conclusions: A data-driven model was able to suggest favorable actions, predict mortality and length of stay with high accuracy. This work provides a solid basis for a scalable probabilistic clinical decision support framework for sepsis treatment that can be expanded to other clinically relevant states and actions, as well as a data-driven model that can be adopted in other clinical areas with sufficient training data.

No MeSH data available.


Related in: MedlinePlus