Sperm Morphology Image Data Set (SMIDS)

SMIDS was collected by a smartphone based data acquisition approach, which has been firstly used in [1] for only motile sperm detection and counting scope. Then, smartphone based this approach has been validated in sperm concentration analysis with high correlation between the manual counting results and the system output [2]. Addition to motile sperm detection/counting and concentration analysis, system was also used for creating a stained sperm image dataset, which named as SMIDS. The full version of SMIDS was firstly introduced in [3] with a hybrid morphological analyzing framework. In SMIDS, the abnormal class is not divided into the sub-classes as in other sperm morphology datasets. Sperm images were labelled as normal (1021), abnormal (1005) and non-sperm (974). The images are stored in RGB color-space. Each image in SMIDS might include noise, multi-sperm head and mixed tails.

[1] Ilhan, H. O., & Aydin, N. (2018). A novel data acquisition and analyzing approach to spermiogram tests. Biomedical Signal Processing and Control, 41, 129-139.
[2] Ilhan, H. O., & Aydin, N. (2020). Smartphone based sperm counting-an alternative way to the visual assessment technique in sperm concentration analysis. Multimedia Tools and Applications, 79(9), 6409-6435.
[3] Ilhan, H. O., Sigirci, I. O., Serbes, G., & Aydin, N. (2020). A fully automated hybrid human sperm detection and classification system based on mobile-net and the performance comparison with conventional methods. Medical & biological engineering & computing, 58(5), 1047-1068.

SMIDS数据集采用基于智能手机的数据采集方法构建，该方法最早在文献[1]中仅用于活动精子的检测与计数范畴。随后，该基于智能手机的方法在精子浓度分析中得到验证，其手动计数结果与系统输出呈现高度相关性[2]。除活动精子检测/计数与浓度分析外，该系统还被用于构建命名为SMIDS的染色精子图像数据集。SMIDS的完整版本首次在文献[3]中被提出，并配套了一套混合形态分析框架。在SMIDS中，异常类并未如其他精子形态学数据集一般进一步划分子类。所有精子图像被标注为正常（1021张）、异常（1005张）与非精子（974张）三类。图像以RGB色彩空间存储，且每张图像均可能包含噪声、多个精子头部以及混杂的尾部结构。

[1] Ilhan, H. O., & Aydin, N. (2018). A novel data acquisition and analyzing approach to spermiogram tests. Biomedical Signal Processing and Control, 41, 129-139.
[2] Ilhan, H. O., & Aydin, N. (2020). Smartphone based sperm counting-an alternative way to the visual assessment technique in sperm concentration analysis. Multimedia Tools and Applications, 79(9), 6409-6435.
[3] Ilhan, H. O., Sigirci, I. O., Serbes, G., & Aydin, N. (2020). A fully automated hybrid human sperm detection and classification system based on mobile-net and the performance comparison with conventional methods. Medical & biological engineering & computing, 58(5), 1047-1068.

搜集汇总

背景与挑战

背景概述

SMIDS是一个精子形态学图像数据集，包含3000张RGB图像，分为正常（1021）、异常（1005）和非精子（974）三类。该数据集通过智能手机采集，适用于精子检测、计数和分类研究，支持图像处理和机器学习应用。