---
title: "Integrating Heterogeneous Echocardiography AI Outputs in Language Space: Learnable Late Fusion with a Supervised LLM"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2610.39464v1"
permalink: "https://feed.astralor.com/items/itm_59c887fc918532c0163170c3"
published: "2026-10-29T06:36:39.000Z"
score: 54
channel: "tech"
tags: ["LLM","Integrating Heterogeneous Echocardiography","Outputs","Language Space","Learnable Late Fusion","Supervised LLM View","View"]
---

# Integrating Heterogeneous Echocardiography AI Outputs in Language Space: Learnable Late Fusion with a Supervised LLM

## At a glance

View- and target-specific echocardiography AI models capture useful but complementary information and may not achieve sufficient predictive performance in isolation. A learned ensemble is a natural way to combine these …

## Key points

- View- and target-specific echocardiography AI models capture useful but complementary information and may not achieve sufficient predictive performance in isolation.
- A learned ensemble is a natural way to combine these outputs, but heterogeneous models typically require each output to be converted into a binary decision, probability, confidence score, or fixed-length feature vector.
- We developed a supervised large language model (LLM)-based late-fusion module that directly aggregates view-specific predicted finding statements and predicted measurements in language space and learns a case-level fusi…

## Full text

View- and target-specific echocardiography AI models capture useful but complementary information and may not achieve sufficient predictive performance in isolation. A learned ensemble is a natural way to combine these outputs, but heterogeneous models typically require each output to be converted into a binary decision, probability, confidence score, or fixed-length feature vector. We developed a supervised large language model (LLM)-based late-fusion module that directly aggregates view-specific predicted finding statements and predicted measurements in language space and learns a case-level fusion function. In patient-level three-fold cross-validation of 859 examinations from 646 patients, the proposed method integrated heterogeneous outputs within a single framework, performed comparably to the best-performing reference strategy evaluated, and outperformed a general-purpose LLM without task-specific training. These results suggest that a supervised LLM can combine multiple upstream models without model-specific output conversion or feature engineering that requires medical knowledge.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_59c887fc918532c0163170c3