What’s different in PATHFx v3.0? A contemporary patient population.
We’ve upgraded PATHFx using more contemporary patients, in an effort to modernize the models. We designed the open source PATHFx tool to better integrate with providers, electronic health records, registries and other industry databases. Our machine-learning code, created in-house specifically for PATHFx, allows for greater compatibility and collaboration, while also enabling us to streamline the evolution of new models.
More data equals greater accuracy.
The new models include nearly two decades of clinical experience gleaned since the development of the original PATHFx tool. We’ve also updated the machine learning algorithms to keep pace with the ever-evolving industry standard, and fine-tuned the imputation strategies to better account for missing data in this unique patient population.
Incorporates new advances.
Staying current is vital, especially given the variety of new treatments available for patients with advanced malignancy, including checkpoint inhibition and adoptive immunotherapy. With this in mind, we updated the PATHFx models by training–and externally validating them in patients treated between 2015 and 2018. In addition, we’ve developed PATHFx v3.0 to be useful in patients who are undergoing palliative radiotherapy, as well as surgery for symptomatic lesions.
The PATHFx Advantage: Personalized Treatment of Metastatic Bone Disease
Like the original PATHFx tool, our new model allows orthopaedic surgeons to more accurately assess a patient’s prognosis and determine the most favorable course of action.
It offers:
- Precision Medicine
- PATHFx converts basic demographics and physiological parameters into a patient-specific survival trajectory useful for surgical decision-making.
Optimal Surgeon Control
No one knows your patients better than you do. We offer surgeons a means to weight estimates–essential for our diverse patient population.
External Validation
The PATHFx models are based on Random Forest and Bayesian statistics and have been Externally validated in several patient populations, worldwide.