Abstract
Till date, histopathological examination of concerned tissue by light microscopy is considered to be the gold standard and most acceptable method for the final diagnosis of disease processes. Sometimes, examination of serial sections, i.e., consecutive sections obtained from a histopathologically processed tissue specimen using a microtome, plays a very vital role in comprehensive understanding of the tissue details, aiding in treatment planning, prognosis, and final diagnosis. In this study, the histopathological dataset showcased, focuses on images of serial sections from colonic and pancreatic tissues, captured through light microscopy. These sequential images might serve as a valuable resource for generating a three-dimensional representation of histological tissue samples. The resulting 3D reconstructed data obtained from the serial sections will provide detailed structural information at a high resolution. Although whole-slide imaging is considered a better option to get images of all sections on one slide at multiple desired magnifications and is obviously a more wanted option for 3D reconstruction of an entire tissue, its high cost poses a significant barrier. In this study, the dataset is prepared and collected from the histopathology division of the Department of Pathology, North Bengal Medical College, near Siliguri. It consisted of 168 serial section images of colon and pancreatic tissue captured at different magnifications. This comprehensive dataset will aid biomedical researchers in the field of histopathology analysis, an area that still holds potential for recent advancements, particularly in 3D reconstruction.
Keywords: Serial section, Histopathology, Light microscope, Microtome, Colon, Pancreas
Specifications Table
| Subject | Biomedical Engineering |
| Specific subject area | Three Dimensional reconstruction from light microscopic serial section histopathology images of colon and Pancreas tissue |
| Data format | Raw data consisting of histopathology images having RGB color in .tif format |
| Type of data | Raw Data |
| Data collection | Histopathology tissue serial section slides of the colon and pancreas were prepared at North Bengal Medical College. 48 serial sections were prepared for colon tissue, and 36 serial sections were prepared for Pancreatic tissue. Ethical approval was obtained from North Bengal Medical College and Hospital. These serial sections are then converted to images using a digital microscope named Olympus with lens types 10× and 40× respectively. For colon tissue, 48 serial section images in 10× resolution and 48 images in 40× resolution were captured. Similarly, for pancreatic tissue 36 images were captured in both 10× and 40× resolution. So, a total of 96 images for colon tissue and 72 for pancreas were captured. |
| Data source location | Institution: North Bengal Medical College City/Town/Region: Siliguri, North Bengal Country: India |
| Data accessibility | Repository name: Mendeley data Data identification number: 10.17632/fzskwpzs5j.1 Direct URL to data: https://data.mendeley.com/datasets/fzskwpzs5j/1 |
1. Value of the Data
-
•
A sequence of sections cut in succession from a prepared specimen using a microtome is known as a serial section. The examination of histological section datasets allows for the visualization of internal topographic structures within a specimen that would otherwise remain unseen, providing valuable insights for pathology research and diagnosis [1,2].
-
•
Understanding the three-dimensional structure of an organ or tissue under a light microscope requires knowledge of serial sections. This dataset can be utilized to create a 3D structure of tissue specimens by overlaying the serial sections, offering comprehensive insights into the cyto-architectural structures of the tissue [3,4].
-
•
Publicly available datasets containing serial section tissue images of the colon and pancreas captured through a light microscope are not found in our course of research. This dataset can be used by researchers in their research on histopathology analysis.
-
•
Whole Slide Imaging (WSI) offers the advantage of digitizing a preselected area or the entire slide at a time. It is also capable of producing 3D structure of an entire tissue sample. However, the expensive scanners, maintenance agreements, IT assistance, digital file storage, and pathologists lack of technological knowledge are some of the limitations of WSI [5,6]. Therefore, it is beneficial to carry out research work in light microscopic serial section images due to its wide availability in all hospitals They are affordably priced and have low maintenance costs.
2. Background
Researchers worldwide are actively engaged in histopathology image analysis using Whole Slide Imaging (WSI) serial section images [4,[7], [8], [9]]. WSI is capable of capturing large tissue areas at a single scan and therefore can generate 3D view from multiple serial sections. However, WSI is expensive and requires specialist knowledge, and therefore hospitals are not able to use it very often.
Diseases of the colon and pancreas, such as colorectal cancer and pancreatic adenocarcinoma, are major health challenges. This dataset focuses on colon and pancreatic serial section tissue images, providing targeted information relevant to diseases and conditions affecting these organs. The primary motivation behind creating this dataset is to utilize the light microscope, which is widely available in all hospitals, to generate 3D volumes from serial sections. Researchers can examine the tissue in a continuous succession by using serial sections, which offer successive slices of the tissue. The dataset can help to provide valuable insights into the histological features that can be gathered from the consecutive sections of the same tissue, assisting pathologists and clinicians in diagnosing and characterizing a range of abnormalities in colon and pancreatic tissues. This dataset can facilitate the creation of 3D volumes by overlaying the 2D serial section images, thereby providing valuable depth information.
3. Data Description
The histopathology imagery dataset consists of serial section images of colonic and pancreatic tissues. The serial section dataset contains 48 colon histopathology images with 10x resolution and another 48 colon images with 40× resolution. It also contains 36 pancreas serial section images in 10x resolution and another 36 images in 40× resolution. Table 1 depicts the number of slides along with the number of serial sections for each tissue sample. All the images are stained with Hematoxylin and Eosin. The dataset is uploaded in the Mendeley data repository. A detailed summary of the dataset location in Mendeley database is given in Fig. 2. In Fig. 1, the first three serial section images that were captured are shown. The remaining images can be found in the Mendeley data repository.
Table 1.
Number of slides provided by the pathologist and the number of images captured.
| Sl. No | Tissue | No. of slides | Sections per slide | Total number of images |
|---|---|---|---|---|
| 1. | Colon | 4 | 12 | 48 |
| 2. | Pancreas | 3 | 12 | 36 |
Fig. 2.
Folder description of Serial Section Dataset in Mendeley database.
Fig. 1.
First three serial sections of colon and pancreas tissue.
4. Experimental Design, Materials and Methods
Histopathology serial section slides were prepared at the Department of Pathology, North Bengal Medical College. Tissue samples were gathered from patients, whose identities have been anonymized for the purpose of this study. The entire process of tissue processing and converting it to digitized images broadly consists of two steps:
-
i)
Preparation of serial section histopathology slides of colon and pancreas.
-
ii)
Capturing the two dimensional images of the serial sections using different magnification of a light microscope.
One of the most important steps in histological or cytological observation is sample and slide preparation. The initial step involves acquiring fresh tissue specimens from the colon and pancreas following a surgical procedure. These biological samples are then placed in a formalin solution for complete fixation in a fixative solution [1]. Since most fresh tissue is fragile and susceptible to distortion and damage, thin slices cannot be prepared from it without chemical preservation, or fixation, to maintain its integrity during the cutting process. Subsequently, the sample undergoes dehydration. The specimen needs to be dried before it can be penetrated with a liquid solution to change it into a solid state suitable for cutting thin slices. Following dehydration, the sample is embedded by solidifying it in a paraffin embedding medium. The specimen is fully penetrated by wax, and is given a shape of a block. In the next step, sectioning is being done using a microtome [1,2]. To reveal the tissue, the block's surface is cleared of wax. Using a microtome, very thin ribbon-shaped tissue slices are cut from the block. Most tissues are sliced at approximately 3–5 µm, however, the microtome can be programmed to cut at other thicknesses. After being cut, the tissue ribbons are gently placed in a bath of warm water, and then they are placed onto a slide. The last step of slide preparation was staining the sections. When unstained, the majority of cells are translucent and nearly colorless. Therefore, tissue sections are stained with standard histochemical stains called hematoxylin and eosin, to give contrast and enhance the visibility and evaluation of tissue structures. The overall steps that were involved in the tissue section preparation is depicted in Fig. 3.
Fig. 3.
Stages of histology slide preparation.
Once the slides are prepared, the next step was to convert the tissue serial sections to two dimensional images. The number of slides provided by the pathologist and the total number of images captured for each class are listed in Table 1 given above.
A total of 168 serial section images of colon and pancreas tissue were captured from raw sample slides using a digital microscope from the Olympus model (BX41 Dual head microscope). The slides containing the serial sections is shown in Fig. 4. Different types of objective lens were used (×10, and ×40) to obtain different levels of magnifying details of the serial section images and the eyepiece was of 10× power. 0.25 and 0.65 was the numerical aperture of the 10× and 40× lens respectively. The images are saved in .tif format having dimension of 1280 × 960.
Fig. 4.
Histology slides of colon and pancreas serial sections.
Limitations
Limited number of serial sections could be produced from the tissue specimen due to very thin and delicate structure of tissues. While capturing images of serial sections in a sequence it is seen that the tissue structures lose similarity and gets distorted and therefore it becomes difficult to identify the same region in the subsequent sections.
Ethics Statement
This data collection did not involve any clinical experiment or direct involvement of patients. No informed consent was obtained specifically for the tissue collection procedure. Patients were verbally informed about the collection of samples during their clinic visits. Ethical approval for data collection was obtained from the North Bengal Medical College, Siliguri with reference number: IEC/NBMC/M-11/019/2024
CRediT authorship contribution statement
Bijoyeta Roy: Conceptualization, Methodology, Software, Writing – original draft. Mousumi Gupta: Data curation, Supervision, Writing – review & editing. Bidyut Krishna Goswami: Visualization, Investigation, Supervision, Writing – review & editing.
Acknowledgments
Acknowledgments
We would like to thank Dr. Madhurima Das, Junior Resident, Department of Pathology, North Bengal Medical College, Siliguri, for providing her valuable feedback and insights in the entire dataset preparation process.
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Data Availability
Serial Section Histopathology Image Dataset (Original data) (Mendeley Data).
References
- 1.Slaoui M., Fiette L. Histopathology procedures: from tissue sampling to histopathological evaluation. Drug Saf. Eval.: Methods Protocols. 2011;691:69–82. doi: 10.1007/978-1-60761-849-2_4. [DOI] [PubMed] [Google Scholar]
- 2.Slaoui M., Bauchet A.L., Fiette L. Tissue sampling and processing for histopathology evaluation. Drug Saf. Eval.: Methods Protocols. 2017;1641:101–114. doi: 10.1007/978-1-4939-7172-5_4. [DOI] [PubMed] [Google Scholar]
- 3.Clarke G.M., Eidt S., Sun L., Mawdsley G., Zubovits J.T., Yaffe M.J. Whole-specimen histopathology: a method to produce whole-mount breast serial sections for 3-D digital histopathology imaging. Histopathology. 2007;50(2):232–242. doi: 10.1111/j.1365-2559.2006.02561.x. [DOI] [PubMed] [Google Scholar]
- 4.Song Y., Treanor D., Bulpitt A.J., Magee D.R. 3D reconstruction of multiple stained histology images. J. Pathol. inform. 2013;4(2):7. doi: 10.4103/2153-3539.109864. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Krishnamurthy S., Mathews K., McClure S., Murray M., Gilcrease M., Albarracin C., Spinosa J., Chang B., Ho J., Holt J., Cohen A., Liang K. Multi-institutional comparison of whole slide digital imaging and optical microscopy for interpretation of hematoxylin-eosin–stained breast tissue sections. Arch. Pathol. Lab. Med. 2013;137(12):1733–1739. doi: 10.5858/arpa.2012-0437-OA. [DOI] [PubMed] [Google Scholar]
- 6.Farahani N., Parwani A.V., Pantanowitz L. Whole slide imaging in pathology: advantages, limitations, and emerging perspectives. Pathol. Lab. Med. Int. 2015:23–33. [Google Scholar]
- 7.Roberts N., Magee D., Song Y., Brabazon K., Shires M., Crellin D., Orsi N.M., Quirke R., Quirke P., Treanor D. Toward routine use of 3D histopathology as a research tool. Am. J. Pathol. 2012;180(5):1835–1842. doi: 10.1016/j.ajpath.2012.01.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kurz A., Müller H., Kather J.N., Schneider L., Bucher T.C., Brinker T.J. 3-dimensional reconstruction from histopathological sections: a systematic review. Lab. Investig. 2024;104(6) doi: 10.1016/j.labinv.2024.102049. [DOI] [PubMed] [Google Scholar]
- 9.Roy M., Wang F., Teodoro G., Bhattarai S., Bhargava M., Rekha T.S., Aneja R., Kong J. Deep learning based registration of serial whole-slide histopathology images in different stains. J. Pathol. Inform. 2023;14:100311. doi: 10.1016/j.jpi.2023.100311. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Serial Section Histopathology Image Dataset (Original data) (Mendeley Data).




