Skip to content

The transition probabilities for 25 transitions from initial state to each intermediate and absorbing states were estimated using a Kaplan-Meier test through the wide-format datset. Transition probabilities for hazard functions were estimated using the lifelines package in Python within a Visual Studio Code integrated development environment.

Notifications You must be signed in to change notification settings

rrwabina/CKD-Multi-State

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 

Repository files navigation

ROMEN SAMUEL RODIS WABINA
THESIS PROGRESS REPORT 

**Date:** 31 December 2024  

## Progress Overview  

### [A] CKD Multi-State Model  
- **Completed Tasks:**  
  - ✅ Univariate Analysis for continuous variables (per transition)  
  - ✅ Univariate Analysis for dichotomous variables (per transition)  
  - ✅ Multi-State wide format  
  - ✅ Multi-State long format  

- **Validation of Multi-State Long Format:**  
  - ✅ Verified censored dates (especially for CVD)  
  - ✅ Verified 25 transitions  

- **Analysis and Model Development:**  
  - ✅ Imputation manuscript (ready for submission)  
  - ✅ Univariate Analysis  
  - ✅ Multivariate Analysis  
  - ✅ Random Survival Forest  
  - ✅ Survival Support Vector Machine (SVM)  
  - ✅ DeepSurv  
  - ✅ Recalibration Techniques:  
    - Isotonic Regression  
    - Platt Scaling  
    - Conformal Prediction  

---

### [B] Time-Varying Imputation  
- **Completed Tasks:**  
  - ✅ Imputation datasets generated for m = 5, 10, 20, 30, 40, 50, 60, 70, 80, 90 (ongoing)  

---


#######################################################################################
Date: 31 December 2024
Progress Report

[A] CKD Multi-State Model 
- DONE: Univariate Analysis for continuous  variables (per transition)
- DONE: Univariate Analysis for dichotomous variables (per transition)
- DONE: Multi-State wide format
- DONE: Multi-State long format 

- [DONE]: Check Multi-State long format
        - [DONE] Verify censored dates (espetcially for CVD)
        - [DONE] Verify 25 transitions

- [DONE]: Imputation manuscript (ready to submit)
- [DONE]: Univariate Analysis 
- [DONE]: Multivariate Analysis 
- [DONE]: Random Survival Forest
- [DONE]: Survival SVM
- [DONE]: DeepSurv
- [DONE]: Recalibration: Isotonic Regression, PLatt Scaling, Conformal Prediction

[B] Time-varying imputation
- [DONE]: Imputation dataset m = 5, 10, 20, 30, 40, 50, 60, 70, 80, 90 (ongoing)

#######################################################################################
Date: 27 October 2024
Progress Report

[A] CKD Multi-State Model 
- DONE: Univariate Analysis for continuous  variables (per transition)
- DONE: Univariate Analysis for dichotomous variables (per transition)
- DONE: Multi-State wide format
- DONE: Multi-State long format 

- [DONE]: Check Multi-State long format
        - [DONE] Verify censored dates (espetcially for CVD)
        - [DONE] Verify 25 transitions

- [DONE]: Imputation manuscript (ready to submit)
- [DONE]: Univariate Analysis 
- [    ]: Multivariate Analysis 
- [DONE]: Random Survival Forest
- [DONE]: Survival SVM
- [DONE]: DeepSurv
- [DONE]: Recalibration: Isotonic Regression, PLatt Scaling, Conformal Prediction

[B] Longitudinal Imputation
- [    ]: Proof: Uncertainty-Aware Non-Stationary State-Space Model with MICE
                - Check covariance matrix proof as uncertainty source
                - Improve expected improvement function as a time-series approach
- [DONE]: Code 
- [    ]: Multivariate Amputation per patient (check how to do it)

#######################################################################################
Date: 25 September 2024
Progress Report

[A] Imputation progress:
ALL MODELS DONE!!!! (28 August 2024)
- DONE: Conventional MICE 
- DONE: Uncertainty-Aware MICE Models
- DONE: RUN BACKUP IMPUTATION MODELS
    
[B] CKD Multi-State Model 
- DONE: Univariate Analysis for continuous  variables (per transition)
- DONE: Univariate Analysis for dichotomous variables (per transition)
- DONE: Multi-State wide format
- DONE: Multi-State long format 

- [DONE]: Check Multi-State long format
        - [DONE] Verify censored dates (espetcially for CVD)
        - [DONE] Verify 25 transitions

- [DONE]: Imputation manuscript (version 2)
        Note: Ajarn Ammarin finished checking first draft (03 September 2024)
        Deadline (draft 2): 15 September 
        [DONE] Introduction
        [DONE] Methodology 
        [DONE] Results and Discussion

#######################################################################################
Date: 11 September 2024
Progress Report

[A] Imputation progress:
ALL MODELS DONE!!!! (28 August 2024)
- DONE: Conventional MICE 
- DONE: Uncertainty-Aware MICE Models
- DONE: RUN BACKUP IMPUTATION MODELS
    
[B] CKD Multi-State Model 
- DONE: Univariate Analysis for continuous  variables (per transition)
- DONE: Univariate Analysis for dichotomous variables (per transition)
- DONE: Multi-State wide format
- DONE: Multi-State long format 

- [DONE]: Check Multi-State long format
        - [DONE] Verify censored dates (especially for CVD)
        - [DONE] Verify 25 transitions

- [  ]: Imputation manuscript (version 2)
        Note: Ajarn Ammarin finished checking first draft (03 September 2024)
        Deadline (draft 2): 15 September 
        [   ] Introduction
        [   ] Methodology 
        [   ] Results and Discussion

#######################################################################################
Date: 10 September 2024
Progress Report

[A] Imputation progress:
ALL MODELS DONE!!!! (28 August 2024)
- DONE: Conventional MICE 
- DONE: Uncertainty-Aware MICE Models
- DONE: RUN BACKUP IMPUTATION MODELS
    

[B] CKD Multi-State Model 
- DONE: Univariate Analysis for continuous  variables (per transition)
- DONE: Univariate Analysis for dichotomous variables (per transition)
- DONE: Multi-State wide format
- DONE: Multi-State long format 

- [DONE]: Check Multi-State long format
        - [DONE] Verify censored dates (especially for CVD)
        - [DONE] Verify 25 transitions

- [  ]: Imputation manuscript (version 2)
        Note: Ajarn Ammarin finished checking first draft (03 September 2024)
        Deadline (draft 2): 15 September 

#######################################################################################
Date: 06 September 2024
Progress Report

[A] Imputation progress:
ALL MODELS DONE!!!! (28 August 2024)
- DONE: Conventional MICE 
- DONE: Uncertainty-Aware MICE Models
- [  ]: RUN BACKUP IMPUTATION MODELS
    

[B] CKD Multi-State Model 
- DONE: Univariate Analysis for continuous  variables (per transition)
- DONE: Univariate Analysis for dichotomous variables (per transition)
- DONE: Multi-State wide format
- DONE: Multi-State long format 

- [  ]: Check Multi-State long format
        - [    ] Verify censored dates (especially for CVD)
        - [DONE] Verify 25 transitions

- [  ]: Imputation manuscript (version 2)
        Note: Ajarn Ammarin finished checking first draft (03 September 2024)
        Deadline (draft 2): 15 September 

#######################################################################################
Date: 15 August 2024
Progress Report

[A] Imputation progress:
- DONE: Conventional MICE 
- DONE: Uncertainty-Aware MICE Models
    [x] Uncertainty-Aware LinearReg (5 datasets)
    [x] Uncertainty-Aware DT (5 datasets)
    [x] Uncertainty-Aware RF (5 datasets)
    [x] Uncertainty-Aware XGBoost (5 datasets)
    [x] Uncertainty-Aware LinearReg (40 datasets)
    [x] Uncertainty-Aware LinearReg (50 datasets)
    [x] Uncertainty-Aware LinearReg (60 datasets)
    [x] Uncertainty-Aware DT (40 datasets)
    [x] Uncertainty-Aware DT (50 datasets)
    [x] Uncertainty-Aware DT (60 datasets)

[B] CKD Multi-State Model 
- DONE: Univariate Analysis for continuous  variables (per transition)
- DONE: Univariate Analysis for dichotomous variables (per transition)
- [  ]: Multi-State wide format
- [  ]: Multi-State long format 

#######################################################################################
Date: 01 August 2024
Progress Report

[A] Imputation progress:
- DONE: Conventional MICE 
- DONE: Uncertainty-Aware MICE Models
    [x] Uncertainty-Aware LinearReg (5 datasets)
    [x] Uncertainty-Aware DT (5 datasets)
    [x] Uncertainty-Aware RF (5 datasets)
    [x] Uncertainty-Aware XGBoost (5 datasets)

[B] CKD Multi-State Model 
- DONE: Univariate Analysis for continuous  variables (per transition)
- DONE: Univariate Analysis for dichotomous variables (per transition)

#######################################################################################
Date: 30 June 2024
Progress Report

[A] CKD Multi-State Model 
- [  ]: Univariate Analysis for continuous  variables (per transition)
- DONE: Univariate Analysis for dichotomous variables (per transition)

[B] CKD IMPUTATION MANUSCRIPT
    [x] Introduction
    [x] Related Literature
    [x] Methodology
    [x] Algorithm
    [x] Proof
    [x] Results (CKD)
    [x] Results (HT)
    [x] Discussion
    [x] Supplementary Information 

#######################################################################################
Date: 15 June 2024
Progress Report

[A] CKD Multi-State Model 
- [  ]: Univariate Analysis for continuous  variables (per transition)
- [  ]: Univariate Analysis for dichotomous variables (per transition)

[B] CKD IMPUTATION MANUSCRIPT
    [x] Introduction
    [x] Related Literature
    [x] Methodology
    [x] Algorithm
    [x] Proof
    [x] Results (CKD)
    [ ] Results (HT)
    [ ] Discussion
    [ ] Supplementary Information 

#######################################################################################
Date: 01 June 2024
Progress Report

[A] CKD IMPUTATION MANUSCRIPT
    [x] Introduction
    [ ] Related Literature
    [x] Methodology
    [x] Algorithm
    [x] Proof
    [ ] Results (CKD)
    [ ] Results (HT)
    [ ] Discussion
    [ ] Supplementary Information 

About

The transition probabilities for 25 transitions from initial state to each intermediate and absorbing states were estimated using a Kaplan-Meier test through the wide-format datset. Transition probabilities for hazard functions were estimated using the lifelines package in Python within a Visual Studio Code integrated development environment.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published