Scholarlink

Background <p>Accurate diagnosis of hematologic malignancies from peripheral blood smears (PBSs) requires integrating cellular morphology and composition across numerous white blood cells. Existing computational approaches predominantly automate single-cell classifications and do not provide holistic, slide-level diagnostic predictions.</p> Methods <p>We present a framework that employs a high-performance cell-based encoder (DeepHeme) for feature extraction, integrated with our weakly supervised, attention-based multiple instance learning (MIL) model, termed CAREMIL (Cell AggRegation, Explainable, Multiple Instance Learning). Through comprehensive evaluations of leading image encoders and MIL architectures, the combination of DeepHeme and CAREMIL demonstrated superior performance on disease classification tasks. CAREMIL functions as a robust aggregation mechanism, consistently outperforming established slide-level MIL methods (gated MIL and Dual-stream MIL Network) across multiple encoder types. The most pronounced performance gains were observed with out-of-domain encoders, including ImageNet-pretrained and open-source pathology foundation models (UNI2 and Virchow2).</p> Results <p>CAREMIL combined with DeepHeme achieves the highest diagnostic accuracy across acute myeloid leukemia (AML), myelodysplastic syndromes (MDS), and hairy cell leukemia (HCL), with AUROCs of 0.999, 0.891, and 0.945, respectively, and successfully identifies AML even in cases with minimal or absent circulating blasts. Attention values assigned by CAREMIL highlight diagnostically relevant cells and reveal disease-specific morphometric patterns, enabling biological interpretability and case-level insights. The framework remains resilient to individual cell misclassifications and does not require explicit cell-level supervision.</p> Conclusions <p>These findings establish CAREMIL as an effective and interpretable MIL framework for hematologic slide diagnosis, extendable to bone marrow aspirates, cytology, and other liquid biopsy specimens, supporting a shift toward quantitative, morphology-informed hematologic diagnostics.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Interpretable multiple instance learning for hematologic diagnosis from peripheral blood smears

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月15日

查看全文
  • Siddharth Singi,
  • Shenghuan Sun,
  • Zhanghan Yin,
  • Riya Gupta,
  • Dylan C. Webb,
  • Khawaja H. Bilal,
  • Deepika Dilip,
  • Linlin Wang,
  • Neeraj Kumar,
  • Swaraj Nanda,
  • Nicolas Sanchez,
  • Jacob G. Van Cleave,
  • Brenda Fried,
  • Sean Paulsen,
  • Ethan S. Yan,
  • Ali Kamali,
  • Argho Sarkar,
  • Allyne Manzo,
  • Jeeyeon Baik,
  • Irem S. Isgor,
  • Cesar Colorado-Jimenez,
  • Anthony Cardillo,
  • Leonardo Boiocchi,
  • Aijazuddin Syed,
  • David Kim,
  • Brie Kezlarian-Sachs,
  • Maly Fenelus,
  • Alexander Chan,
  • Mariko Yabe,
  • Samuel I. McCash,
  • Menglei Zhu,
  • Simon Mantha,
  • Orly Ardon,
  • Lauren McVoy,
  • Wenbin Xiao,
  • Mikhail Roshal,
  • Oscar Lin,
  • Ahmet Dogan,
  • Iain Carmichael,
  • Chad Vanderbilt,
  • Gregory M. Goldgof

摘要

Background

Accurate diagnosis of hematologic malignancies from peripheral blood smears (PBSs) requires integrating cellular morphology and composition across numerous white blood cells. Existing computational approaches predominantly automate single-cell classifications and do not provide holistic, slide-level diagnostic predictions.

Methods

We present a framework that employs a high-performance cell-based encoder (DeepHeme) for feature extraction, integrated with our weakly supervised, attention-based multiple instance learning (MIL) model, termed CAREMIL (Cell AggRegation, Explainable, Multiple Instance Learning). Through comprehensive evaluations of leading image encoders and MIL architectures, the combination of DeepHeme and CAREMIL demonstrated superior performance on disease classification tasks. CAREMIL functions as a robust aggregation mechanism, consistently outperforming established slide-level MIL methods (gated MIL and Dual-stream MIL Network) across multiple encoder types. The most pronounced performance gains were observed with out-of-domain encoders, including ImageNet-pretrained and open-source pathology foundation models (UNI2 and Virchow2).

Results

CAREMIL combined with DeepHeme achieves the highest diagnostic accuracy across acute myeloid leukemia (AML), myelodysplastic syndromes (MDS), and hairy cell leukemia (HCL), with AUROCs of 0.999, 0.891, and 0.945, respectively, and successfully identifies AML even in cases with minimal or absent circulating blasts. Attention values assigned by CAREMIL highlight diagnostically relevant cells and reveal disease-specific morphometric patterns, enabling biological interpretability and case-level insights. The framework remains resilient to individual cell misclassifications and does not require explicit cell-level supervision.

Conclusions

These findings establish CAREMIL as an effective and interpretable MIL framework for hematologic slide diagnosis, extendable to bone marrow aspirates, cytology, and other liquid biopsy specimens, supporting a shift toward quantitative, morphology-informed hematologic diagnostics.