---
type: paper
slug: suhyeon-kim-2022-rsae-biological-age
title: >-
  Risk score-embedded deep learning for biological age estimation: Development
  and validation
authors:
  - '[[suhyeon-kim]]'
  - '[[hangyeol-kim]]'
  - '[[eun-sol-lee]]'
  - '[[chiehyeon-lim]]'
  - '[[junghyelee]]'
year: 2022
venue: '[[information-sciences]]'
volume: '586'
pages: 628-643
doi: 10.1016/j.ins.2021.12.015
categories:
  - '[[medical-ai]]'
  - '[[machine-learning]]'
methods:
  - '[[autoencoder]]'
  - '[[risk-score-embedding]]'
  - '[[representation-learning]]'
  - '[[kl-divergence]]'
  - '[[t-scale-transformation]]'
themes:
  - '[[biological-age]]'
  - '[[health-index]]'
  - '[[disease-risk-prediction]]'
  - '[[knhanes]]'
  - '[[nhic-cohort]]'
  - '[[hypertension-prediction]]'
  - '[[type-2-diabetes-prediction]]'
aliases:
  - RSAE-BA
last_updated: '2026-05-17T21:30:00+09:00'
---
# Risk score-embedded deep learning for biological age estimation: Development and validation

*[[suhyeon-kim]], [[hangyeol-kim]], [[eun-sol-lee]], [[chiehyeon-lim]], [[junghyelee]] (2022)* · [[information-sciences]] 586:628-643 · [DOI ↗](https://doi.org/10.1016/j.ins.2021.12.015)

> **RSAE-BA (Risk Score-Embedded Autoencoder for Biological Age)** — *health risk* 정보를 [[autoencoder]] embedding 의 loss function 에 *명시 통합* 한 *biological age (BA) 추정* 새 알고리즘. 기존 BA 추정의 3 한계 극복: (1) Linear regression (MLR, lasso, elastic net) — BA 를 CA (chronological age) 로 distort, (2) Deep learning regression (Cole et al 2017 CNN, Pyrkov et al 2018) — CA 를 target 으로 supervised — *BA goal 과 deviation*, (3) Klemera-Doubal Method (KDM 2006) + PCA — assumption-heavy or *variance-maximization direction* 의 *health-irrelevance*. 4 단계: (i) Cumulative Gaussian probability 기반 *Risk Score (RS)* 계산 — Type 1 (large=poor), Type 2 (small=poor), Type 3 (both=poor); (ii) Data transformation 으로 *방향 정렬*; (iii) RSAE modeling — loss $\mathcal{L}_{RSAE} = \mathcal{L}_{KL}(\mathbf{X}', \mathbf{X}'') + \mathcal{L}_{KL}(\mathbf{X}'', \mathbf{P})$; (iv) BA = $T$-scale 변환된 standardized BAS. **새 BA validation** — LR vs HR group 의 RS 차이로 *unlabeled + labeled* 모두 평가. **3 dataset 검증**: KNHANES (85,490명, unlabeled), NHIC-HTN (140,867명, 2.5 년 hypertension 발생), NHIC-T2DM (T2DM 발생). [[junghyelee]] 의 *3 기 TEMEP 시기 의료 AI 정점*.

- **RQ**: (1) BA 추정에 *health risk* 정보를 deep representation learning embedding 에 통합할 수 있는가? (2) BA validation 의 *cohort dependency* 한계 (mortality/disease incidence 필요) 를 어떻게 *unlabeled data* 에도 적용 가능하게 만드는가? (3) RSAE-BA 가 기존 방법 (MLR, lasso, KDM, PCA, vanilla AE) 대비 *질병 incidence 예측* 에 우위인가?
- **방법론**: **Stage 1** — Risk Score (RS) calculation: 각 변수의 normal 분포 $\mathcal{N}(\mu_j, \sigma_j^2)$ ($\mu_j$ = medical prior, $\sigma_j$ = empirical) 의 cumulative probability. 3 type: Type 1 (one-sided large), Type 2 (one-sided small), Type 3 (two-sided). **Stage 2** — Data transformation (Algorithm 1): standardize → Type-specific direction alignment (×-1 or |·|) → sigmoid 로 [0,1] bounding. **Stage 3** — RSAE modeling (Algorithm 2): encoder $\mathbf{H}_{Enc}^{(k)} = f_{ReLU}(\mathbf{H}_{Enc}^{(k-1)} \mathbf{W}_{Enc}^{(k)} + \mathbf{b}_{Enc}^{(k)})$, decoder mirror. KL divergence loss = reconstruction + RS-prediction. **Stage 4** — BA calculation: BAS = $z$ 의 평균 standardize, BA = $T$-scale transformation. **BA validation** — Group 1 (BA<CA) 의 LR group ($q\%$ largest CA−BA) vs Group 2 (BA>CA) 의 HR group ($q\%$ largest BA−CA); hypothesis: $RS_{LR} < RS_{HR}$, *good BA* = larger RS gap
- **데이터**: (i) **KNHANES** (Korea National Health and Nutrition Examination Survey, 1988-2017): 85,490 명 — unlabeled. (ii) **NHIC-HTN** (Korean National Health Insurance Corporation 2002-2010, hypertension cohort): 140,867 명 — 2.5 yr HTN 발생 labeled. (iii) **NHIC-T2DM** (type-2 diabetes cohort): T2DM 발생 labeled. *2% population sample stratified by age, sex, income*. Variables: clinical biomarker (BP, glucose, cholesterol, BMI 등)
- **주요 발견**: (i) **RSAE-BA > MLR / lasso / KDM / PCA-BA / vanilla AE-BA** 에서 *모든 데이터셋* (KNHANES, NHIC-HTN, NHIC-T2DM) 의 *LR vs HR group RS gap* 최대. (ii) **Disease incidence 예측** — RSAE-BA 가 HTN, T2DM 발생 의 *AUC/AUROC* baseline 대비 향상. (iii) **Unlabeled data 검증 가능** — RS 기반 validation 이 *cohort dependency 한계* 해소. (iv) **3 Type RS classification** 의 정합성 — 모든 clinical variable 의 *risk direction* 명시. (v) Sample size 85k → 140k 의 *scalability* 검증
- **시사점**: (a) **Hyper-personalized health management** 의 *quantitative anchor* — BA 를 *useful health index* 로 일상 활용. (b) **RS embedding pattern 의 generalizability** — 다른 health index (e.g. mental health score), 다른 domain (e.g. *household financial risk* of [[suhyeon-kim-2023-household-financial-rihae]]) 로 확장. (c) **Cohort-independent validation** — 일반 EMR (electronic medical record) 만으로 BA index 평가 가능. (d) **Medical prior + data-driven hybrid** — Gaussian $\mu_j$ 의 *medical normal criterion* + $\sigma_j$ 의 *empirical* 결합. (e) 예방 의학 + 만성질환 (HTN, T2DM) 의 early warning index

![Fig. 3 의 RSAE 구조: input $\mathbf{X}'$ → encoder ($K$ layers, ReLU) → bottleneck $\mathbf{Z}$ → decoder → reconstruction $\mathbf{X}''$. Loss $\mathcal{L}_{RSAE} = \mathcal{L}_{KL}(\mathbf{X}', \mathbf{X}'') + \mathcal{L}_{KL}(\mathbf{X}'', \mathbf{P})$ 의 *dual term* 이 $\mathbf{Z}$ 에 *reconstruction information + health risk information* 동시 인코딩. RS $\mathbf{P}$ 가 latent space 에 explicit embedding.](/papers/202203_KimS_Risk_score_embedded_deep_learning_for_biological_age_estimation_Development_and_validation/fig3.png)

## 요약

본 paper 는 [[junghyelee]] 의 *3 기 SNU TEMEP 부임 초기 (2022 published, 2021 work)* 의 *의료 AI 정점*. [[suhyeon-kim]] (W2V-LSA, LBC 의 동일 제 1 저자) + Hangyeol Kim + Eun-Sol Lee + UNIST [[chiehyeon-lim]] (Department of Industrial Engineering & Graduate School of Artificial Intelligence, Bio-Age Inc 의 medical institute affiliation) + [[junghyelee]] (corresponding) 의 *UNIST-Bio-Age industry-academia* 협업. 동기는 *Chronological Age (CA) 의 한계* — 시간 흐름의 단순 측정으로 *functional, structural body capacity + aging* 평가 부정확. *Biological Age (BA)* 가 대안 health index — *EMR 요약*, *연령-의존 biomarker* 결합. 기존 3 type BA 의 *systematic 한계*:

1. **Regression for CA-based** (MLR, lasso, elastic net + CNN deep learning) — *BA-to-CA distortion* + CA 가 target 인 *supervised setting* 의 deviation
2. **Simulation-based** (KDM, Klemera-Doubal 2006) — multidimensional space 의 regression line - biomarker point 거리 minimize — *assumption-heavy*, mortality 예측 우월하지만 data conformity 의존
3. **PCA-based latent feature** — variance-maximize 방향이 *health-relevance* 와 무관

또한 모든 기존 방법이 *risk direction* (variable 의 *bad level* 이 large vs small 인지) 무시. *Validation* 도 *cohort data 의 mortality/incidence label* 의존 — 일반 EMR 적용 어려움. 본 paper 는 *RS-embedded representation learning* 로 두 한계 동시 해결.

**Methodology Stage 1 — Risk Score (RS) calculation.** 각 변수 $\mathbf{x}_j$ 의 normal 분포 $\mathcal{N}(\mu_j, \sigma_j^2)$ — $\mu_j$ = medical examination normal criterion (prior knowledge), $\sigma_j^2$ = data empirical. RS = cumulative Gaussian probability. 3 type:
- **Type 1**: large = poor (예: glucose, BP). One-sided right.
- **Type 2**: small = poor (예: HDL cholesterol). One-sided left.
- **Type 3**: large or small = poor (예: heart rate). Two-sided.

다른 분포 (binomial, multinomial) 로 extensible.

**Stage 2 — Data transformation (Algorithm 1).** standardize → Type-specific direction alignment (Type 2: $\times -1$; Type 3: $|\cdot|$) → sigmoid $s(a) = 1/(1+e^{-a})$ 로 $[0,1]$ bounding. Direction 정렬이 *loss accumulation error* 방지.

**Stage 3 — RSAE modeling (Algorithm 2).** $K$ hidden layer encoder + mirror decoder, ReLU activation. Loss:

$$
\mathcal{L}_{RSAE} = \mathcal{L}_{KL}(\mathbf{X}', \mathbf{X}'') + \mathcal{L}_{KL}(\mathbf{X}'', \mathbf{P})
$$

여기서 $\mathbf{X}'$ = transformed input, $\mathbf{X}''$ = reconstruction, $\mathbf{P}$ = RS matrix. KL divergence 가 $\mathbf{X}''$ 를 reconstruction *+* RS 양쪽에 fit. *Multi-task learning* 이 bottleneck $\mathbf{Z}$ 에 *generic latent + health risk* 동시 인코딩.

**Stage 4 — BA calculation.** BA Score (BAS) = $z_d$ 의 평균 standardize:

$$
\text{BAS} = \frac{1}{D^{(K)}} \sum_{d=1}^{D^{(K)}} (\mathbf{z}_d - z_d^\mu \mathbf{1}_n) \oslash (z_d^\sigma \mathbf{1}_n)
$$

BA = $T$-scale transformation (standard score → $T$-score) 으로 *year scale* 의 BA.

**BA validation (novel)**. 개인 sort: Group 1 (BA<CA), Group 2 (BA>CA). LR (low-risk) group = Group 1 의 $q\%$ largest CA−BA. HR (high-risk) group = Group 2 의 $q\%$ largest BA−CA. 가설: (i) $RS_{LR} < RS_{HR}$, (ii) good BA = *larger RS gap*. *Unlabeled + labeled* 모두 적용 가능 — *cohort 의존성 해소*.

**Experiments** — 3 dataset:
- **KNHANES** (85,490명, unlabeled): RS-based validation
- **NHIC-HTN** (140,867명, hypertension cohort): 2.5 yr HTN 발생 prediction
- **NHIC-T2DM**: type-2 diabetes 발생 prediction

결과: RSAE-BA 가 모든 dataset 에서 baseline (MLR-BA, lasso-BA, KDM, PCA-BA, vanilla AE-BA) 대비 *LR-HR RS gap 최대* + *disease incidence 예측 AUROC 최대*. 한계: (i) Gaussian normal 가정 의 *all variable applicability*, (ii) $q\%$ 의 ad-hoc selection, (iii) 한국 cohort 한정 — 다른 인구 generalizability 미검증, (iv) hyperparameter ($K$, $D^{(K)}$, learning rate) tuning data-dependent. [[junghyelee]] author page 의 *3 기 의료 AI* 라인의 *peak* + *suhyeon-kim 4-paper series* 의 *medical AI representative*.

## 핵심 결과

**3 dataset 검증**

| Dataset | Sample | Type | 사용 |
|---|---:|---|---|
| **KNHANES** | 85,490 | Unlabeled (annual survey 1988-2017) | RS-based BA validation |
| **NHIC-HTN** | 140,867 | Labeled (2.5 yr HTN 발생) | Hypertension prediction |
| **NHIC-T2DM** | (similar) | Labeled (T2DM 발생) | Type-2 diabetes prediction |

**RS 3 type classification**

| Type | Direction | Distribution | 예 |
|---|---|---|---|
| 1 | Large = poor | One-sided (right) | Glucose, Blood Pressure |
| 2 | Small = poor | One-sided (left) | HDL Cholesterol |
| 3 | Both = poor | Two-sided | Heart rate, BMI |

**비교 baseline (Section 3)**

| Method | 약점 |
|---|---|
| MLR-BA | CA distortion |
| Lasso-BA, Elastic Net-BA | 동일 |
| KDM (Klemera-Doubal) | Assumption-heavy |
| PCA-BA | Variance direction ≠ health |
| Vanilla AE-BA | Generic latent, no health risk |
| **RSAE-BA** | **모든 결함 해결** |

> **정량 결론.** RSAE-BA 가 모든 데이터셋에서 *LR-HR RS gap 최대* + *disease incidence AUROC 최대*. *RS-embedded loss* 의 dual-term 이 latent space 에 health risk 명시 인코딩. Cohort-independent validation 가능.

## 방법론 노트

**Risk Score**: 변수 $\mathbf{x}_j$ 의 RS $\mathbf{p}_j \in [0,1]^n$ = Gaussian $\mathcal{N}(\mu_j, \sigma_j^2)$ 의 cumulative probability. Type 별 one-sided 또는 two-sided.

**RSAE encoder** (Eq 2):

$$
\mathbf{H}_{Enc}^{(k)} = f_{ReLU}(\mathbf{H}_{Enc}^{(k-1)} \mathbf{W}_{Enc}^{(k)} + \mathbf{1}_n \cdot (\mathbf{b}_{Enc}^{(k)})^T)
$$

**RSAE loss** (Eq 6) — KL divergence dual term:

$$
\mathcal{L}_{RSAE} = \mathcal{L}_{KL}(\mathbf{X}', \mathbf{X}'') + \mathcal{L}_{KL}(\mathbf{X}'', \mathbf{P})
$$

여기서 $\mathcal{L}_{KL}(x, y) = \sum_i x_i \log(x_i / y_i)$.

**BA calculation** (Eq 7):

$$
\text{BAS} = \frac{1}{D^{(K)}} \sum_{d=1}^{D^{(K)}} (\mathbf{z}_d - z_d^{\mu} \mathbf{1}_n) \oslash (z_d^{\sigma} \mathbf{1}_n)
$$

$T$-scale: $T_i = 10 \cdot \text{BAS}_i + 50$ (또는 CA 와 align). BA in years.

**BA validation**:

$$
\text{LR} = \{i \in \text{Group 1} \mid \text{CA}_i - \text{BA}_i \in \text{top } q\%\}
$$

$$
\text{HR} = \{i \in \text{Group 2} \mid \text{BA}_i - \text{CA}_i \in \text{top } q\%\}
$$

Test: $\bar{RS}_{LR} < \bar{RS}_{HR}$ + gap 의 크기.

식별은 (i) 85k-140k sample 의 large-scale variation, (ii) 3 dataset 의 *unlabeled + labeled* 동시 검증, (iii) baseline 다양성 (linear, simulation, latent feature, vanilla AE), (iv) RS embedding 의 *interpretable* 의미 (Gaussian + medical prior). *Limitation*: Gaussian 가정 + ad-hoc $q$ + 한국 cohort 한정.

## 연구 계보

본 paper 의 *BA estimation* lineage: Hannum et al (2013) DNA methylation BA, Levine et al (2018) *PhenoAge*, Klemera-Doubal (2006 *Mechanisms of Aging and Development*, KDM), Cole et al (2017 *NeuroImage*, CNN brain age), Pyrkov et al (2018 *Scientific Reports*), Rahman-Adjeroh (2019). *Health index* lineage: Mitnitski-Mogilner-Rockwood (2001 frailty index), DeAngelis et al (2009). *Autoencoder*: Hinton-Salakhutdinov (2006 *Science*), Bengio et al (2013 *PAMI*), Vincent et al (2010 denoising AE). *Risk Score 의 cumulative Gaussian*: medical statistics 전통 — Z-score, percentile rank. *KL divergence*: Kullback-Leibler (1951).

TEMEP 내 sibling: [[suhyeon-kim]] 의 *4-paper series* 의 *medical AI peak* — [[suhyeon-kim-2020-w2v-lsa-blockchain]] (text embedding) → [[suhyeon-kim-2021-lbc-data-mining]] (logistics) → 본 paper (medical AI) → [[suhyeon-kim-2025-tmf-gnn]] (graph neural network). 직접 후속: *RI-HAE* (Risk-Information Hybrid Autoencoder) 의 *household financial risk* domain extension — RSAE 의 architecture 가 medical + financial 도메인 모두 transferable. [[junghyelee]] author page 의 *실타래 3 (Healthcare AI) + 실타래 4 (Representation Learning) 합류* 의 anchor. [[chiehyeon-lim]] (UNIST AI graduate school) 과의 협업 — UNIST 의 *Bio-Age Inc 산학 cluster* 의 academic output.

## See also

- [[autoencoder]]
- [[risk-score-embedding]]
- [[representation-learning]]
- [[kl-divergence]]
- [[t-scale-transformation]]
- [[biological-age]]
- [[health-index]]
- [[disease-risk-prediction]]
- [[knhanes]]
- [[nhic-cohort]]
- [[hypertension-prediction]]
- [[type-2-diabetes-prediction]]
- [[junghyelee]]
- [[suhyeon-kim]]
- [[chiehyeon-lim]]
- [[hangyeol-kim]]
- [[suhyeon-kim-2020-w2v-lsa-blockchain]]
- [[suhyeon-kim-2021-lbc-data-mining]]
- [[suhyeon-kim-2025-tmf-gnn]]
- [[information-sciences]]