Clinical Reasoning Under a Partially Observed Objective in Cone Beam CT Report Generation

基于部分观测目标的锥形束CT报告生成中的临床推理

Abstract: Maxillofacial report generation from cone beam computed tomography is scored here by a composite objective placing 80% of its weight on a large language model judgement of factual entailment and 20% on lexical overlap, of which only the lexical fifth is visible during development.

摘要： 本文针对锥形束计算机断层扫描（CBCT）的颌面部报告生成，提出了一种复合评分目标。该目标将80%的权重分配给大语言模型对事实蕴含的判断，其余20%分配给词汇重叠度，其中在开发阶段仅有词汇重叠部分是可见的。

The grader’s BLEU-4 and METEOR routines are reproduced in pure Python and match the reference to machine precision, and an offline entailment surrogate, which tells a report written for one patient from one written for another at an area under the curve of 0.987, makes the composite objective cheap enough to optimise directly.

评分器的 BLEU-4 和 METEOR 程序均使用纯 Python 重现，并与参考实现达到了机器精度的一致。此外，一个离线蕴含代理模型（能够以 0.987 的曲线下面积区分不同患者的报告）使得该复合目标在计算上足够高效，从而可以直接进行优化。

Over the 622-case public release, a report selected against the visible lexical ranking scores 0.2909, whereas one selected against the composite objective scores 0.4122, because pursuing n-gram overlap drives entailment precision from 0.522 down to 0.266.

在 622 个案例的公开数据集上，基于可见词汇排名选择的报告得分为 0.2909，而基于复合目标选择的报告得分为 0.4122；这是因为单纯追求 n-gram 重叠会导致蕴含准确率从 0.522 下降至 0.266。

A 29 million parameter encoder fine-tuned on the release reaches a prevalence-weighted out-of-fold area under the curve of 0.486 over 985 statements, indistinguishable from the corpus prior, while nine numbers read from the image header reach 0.945 for mandible coverage and 0.872 for condyle coverage, and acquisition centre alone predicts sentence choice at 0.718 against 0.663 for the image-derived model, identifying dictation convention rather than anatomy as the quantity the lexical metrics reward.

一个在发布数据集上微调的 2900 万参数编码器，在 985 条陈述上的患病率加权交叉验证曲线下面积（AUC）为 0.486，与语料库先验无异。与此同时，从图像头文件中读取的九个数值在下颌覆盖率上达到了 0.945，在髁突覆盖率上达到了 0.872；仅凭采集中心这一项指标对句子选择的预测准确率就达到了 0.718，而基于图像的模型仅为 0.663，这表明词汇指标所奖励的实际上是口述习惯，而非解剖结构。

The delivered system emits eight unconditional statements and five gated on header geometry under polarity, laterality and tooth-level consistency constraints, and reaches METEOR 0.3542 over 50 held-out cases from an unseen centre.

最终交付的系统在极性、侧向性和牙位一致性约束下，能够输出八条无条件陈述和五条基于头文件几何信息的门控陈述，并在来自未见中心的 50 个留出案例上达到了 0.3542 的 METEOR 分数。