卓上顕微鏡に代わる病理組織学画像教育のためのスマートフォン撮影技術

Abstract

In recent years, learning methods have become more diverse, with an increasing number of universities moving away from microscopic sketches to digital imaging methods, including smartphone images and virtual slides. We took photos of hematoxylin–eosin-stained specimens using a smartphone camera placed over the ocular lens of an optical microscope at different distances under different conditions to investigate whether organs could be identified from these photos. At an optical microscope magnification of 100× (low-power magnification), in-focus images were obtained without camera shake at smartphone camera magnifications of ×1.0 to ×5.0. Images of similar quality to optical micrographs were obtained, allowing organs to be identified by revealing tissue structures and intranuclear and cytoplasmic features. The distance from the smartphone camera to the ocular lens increased as the smartphone camera magnification increased. The best fixation of the smartphone camera was achieved at up to ×5.0. The present results suggest that improvements in the imaging capability of smartphone cameras fitted with a microscopic adapter could lead to more precise imaging.

Translated Abstract

近年，学習方法は多様化しており，顕微鏡スケッチからスマートフォン画像や仮想スライドなどのデジタルイメージング手法に移行する大学が増えています。ヘマトキシリン－エオジン染色標本を，光学顕微鏡の接眼レンズにスマートフォンのカメラを置き，距離を変えて異なる条件で撮影し，これらの写真から臓器が識別できるかどうかを検討した。光学顕微鏡倍率×100（低倍率）では，スマートフォンのカメラ倍率×1.0～×5.0で手ブレのないピント合画像が得られた。光学顕微鏡写真と同様の品質の画像が得られ，組織構造と核内および細胞質の特徴を明らかにすることによって臓器を特定することができた。スマートフォンのカメラから眼レンズまでの距離は，スマートフォンのカメラの倍率が大きくなるにつれて増加した。スマートフォンカメラの最高の固定は最大×4であった。

I  Introduction

At many universities in Japan, training in histopathology as part of the medical education involves sketching tissue-stained specimens examined through a microscope. However, learning methods have diversified in recent years, with a growing number of universities moving from microscope sketching to methods using smartphone images and digital images, including virtual slides.¹⁾ We therefore investigated whether hematoxylin and eosin; HE-stained specimens can be diagnosed using a smartphone camera fitted with a microscope adapter in lieu of a tabletop optical microscope.

II  Materials and Methods

1.  Microscope

1) lympus BX53

2) Ocular lens 10×

3) Objective lens 10×, 40×

2.  Smartphone (Figure 1)

Figure 1 Smartphone used in the study

The iPhone 11 Pro has three lenses (ultra-wide, wide, telephoto) and automatic image focusing. The F-number can be set from 1.8 to 2.4. The camera has a resolution of 12 megapixels.

1) iPhone 11 Pro (Apple)

2) Resolution: 12 megapixels

3) F-number: Ultra-wide: 2.4, Wide: 1.8, Telephoto: 2.0

4) Magnification can capture images at ×0.5 to ×10.0

3.  Organs

Esophagus, stomach, liver, pancreas, thyroid, cerebellum, heart, and lungs, sliced at 2 μm and stained with HE.

4.  Measuring the distance from smartphone to ocular lens

The smartphone was anchored to the ocular lens by holding the smartphone with both hands, holding the ocular lens of the optical microscope with the little finger, ring finger, and middle finger of each hand, and holding the smartphone camera in place using the index finger and thumb. This allowed the right thumb to be free. When the optimal image appeared on the smartphone screen, the shutter function was pressed with the right thumb. If the smartphone was not held in place securely enough, the image would be out of focus because of camera shake. After clicking the smartphone camera icon and selecting ‘Photo’, the smartphone was placed against the ocular lens of an optical microscope with the camera magnification set to ×1.0, × 2.0, ×3.0, ×4.0, ×5.0, × 6.0, or × 7.0 (Figure 2). The distance between the camera and ocular lens was measured when the image was in focus at each magnification. Esophagus was used as the HE-stained specimen, and images were captured with the optical microscope magnification set at ×100 (low-power magnification) and ×400 (high-power magnification).

Figure 2 Method of setting smartphone over the ocular lens

HE-stained specimens of each organ are set in the optical microscope and focused. After clicking the smartphone camera icon and selecting ‘Photo’, the specimens are photographed.

5.  Optical micrography of HE-stained specimens

Rat esophagus, stomach, liver, pancreas, thyroid, cerebellum, heart, and lungs were fixed in 10% neutral buffered formalin (pH 7.4) for 24 h, then embedded in paraffin blocks, sectioned (2 μm), and stained with HE. The mean of five measurements per magnification was used for analysis.

6.  Educational effect

One method used to understand tissue structure with conventional microscopic observations is sketches. Another method is to use a microscope in combination with a smartphone to save images of the examined tissue on the smartphone for later viewing. We administered a descriptive questionnaire survey to clarify differences between these two methods. Participants comprised 55 students enrolled at the university just prior to taking the national exam. The questionnaire comprised four questions. Question 1 asked about the usefulness of the methods in understanding the structure of organ tissues. Question 2 asked about the usefulness of the methods in preparing for the national examination. Question 3 asked about the usefulness of the methods when sketching. Question 4 asked about the usefulness of the methods in aiding memory retention. Responses were given on a 3-point scale, as follows: A) very useful; B) somewhat useful; and C) not useful. A positive response was considered to be a response of A or B.

III  Results

1.  Distance from smartphone to ocular lens

With the esophageal HE-stained specimen focused under the optical microscope at ×100 (low power), the distance between the ocular lens and smartphone camera increased as the camera magnification increased, from 2.1 cm at ×1.0, to 2.3 cm at ×2.0, and 3.5 cm at ×6.0, at which point camera shake resulted in an inability to obtain an in-focus image. At ×400 (high power), the distance between the camera and ocular lens was 2.2 cm when camera magnification was ×1.0, and the distance between the two points similarly increased with each increase in magnification, as with ×100 (low power). At ×7.0, the image was out of focus because of camera shake (Figure 3).

Figure 3 Smartphone photography conditions

Distance between the two points increases with higher camera magnification. Images are in focus up to ×5.0 and ×6.0, respectively, with the optical microscope at ×100 (low power) and ×400 (high power). Images show esophageal HE-stained specimens.

2.  Optical micrography of HE-stained specimens

Images were captured using an Olympus BX53 optical microscope with cellSens imaging software (Olympus Corporation, Tokyo, Japan). All organs were photographed at ×100 (low power) and ×400 (high power) and compared with smartphone photos.

3.  Comparison of HE-stained specimens photographed with smartphone and optical microscope

When the optical microscope was set to ×100 (low magnification), the following cellular structures could also be captured as clear autofocus images at smartphone camera magnifications of ×1.0, ×3.0 and ×5.0. The organs were (Olympus Corporation, Tokyo, Japan). esophageal squamous cells, glandular structures of the stomach, intestinal epithelialization, central veins of the liver, hepatocytes, pancreatic duct and adenocytes, thyroid colloid and follicular cells, molecular layer, granular layer, Purkinje cells in the cerebellum, rhabdomyocytes in the heart, alveolar epithelium and alveolar structures of the lung.

4.  Learning benefits for students

Survey results for the 55 students showed that 85% of respondents answered A or B to Question 1, 76% answered A or B to Question 2, 89% answered A or B to Question 3, and 81% answered A or B to Question 4.

IV  Discussion

Teaching methods using digital images are now being employed in a range of fields, including pathology, hematology, and veterinary science. Smartphones also look set to be embraced for teaching purposes¹⁾ in Japan and around the world. In Japan, smartphone ownership is above 80% and is particularly high among young people. The performance of smartphone cameras has also evolved, and the camera used for this study probably offers the best digital images among current mobile devices, having 12 megapixels and an F-number of 2.0.^2),3) The camera can magnify from ×0.5 to ×10.0, has two different wide-angle lenses and a telephoto lens, with autofocus and automatic lens selection functions. The distance measured from the smartphone camera to the ocular lens increased as the magnification increased. For esophageal HE-stained specimens with the optical microscope at ×100 (low power), the smartphone camera distances at ×1.0 and ×2.0 were 2.1 cm and 2.3 cm, respectively, and the best images were obtained at smartphone magnification up to ×5.0 and distance 3.0 cm. At ×6.0 magnification and higher, the distance between the ocular lens and smartphone camera was too great for the smartphone camera to be held stably enough, leading to camera shake and an inability to obtain in-focus images.⁴⁾ We considered that the best method of fixing the smartphone in place over the ocular lens for depicting optimal images was to hold the smartphone with both hands, anchoring the little, ring, and middle fingers of each hand against the ocular lens and holding the smartphone camera in place using the index finger and thumb. If images can be accurately depicted at low-power magnification (×100) while focusing the smartphone camera at ×1.0, these images can be magnified with a pinch out operation after capture, allowing a better understanding of detailed cell morphology. At high-power magnification (×400), images up to 6.0× could also be obtained. When comparing smartphone photos and optical micrographs, the in-focus smartphone images of specimens from the esophagus, stomach, cerebellum, kidney, and liver taken at low power (×100) and high power (×400) revealed the nuclear chromatin structure stained with hematoxylin and the cytoplasm stained with eosin, thus allowing the organs to be identified. Smartphone photography of this sophistication applied to histopathology training would be of high educational value. Smartphone images can also be magnified by pinch out operation, allowing assessment of the mucosal surface of the digestive tract, hepatocytes, blood vessels, individual cells, etc., and allowing the viewer to grasp cell morphology in greater detail.⁵⁾ However, when capturing microscopic images with a smartphone, increasing the smartphone’s magnification can have adverse effects, such as inability to focus. The best way to capture an image is to take a picture at ×1.0 magnification and then pinch out the image to enlarge it.

Students could also study at home using photos taken on their smartphones through an optical microscope after searching for characteristic cells in HE-stained specimens that they have prepared in histopathology training. The use of a smartphone camera to photograph HE-stained specimens turns these specimens into a more portable resource, and this could enhance the educational effectiveness of training in histopathology. These photographic techniques can be learned quickly with practice, even by beginners, and would allow students to study pathology inexpensively, anytime, and anywhere.⁶⁾ Students also felt that their comprehension improved with the use of the smartphone method. If the use of smartphones does not lead to the improvement of educational effects, we must try to make students more interested, such as incorporating game nature. Smartphone photography is a tool that can be applied to distance learning, patient imaging, and education, and is likely to be utilized in the teaching of cell morphology by more educational facilities in the future.^7),8)

Conclusions

Smartphone cameras can enhance the educational effectiveness of histopathology training and can also be applied to telediagnosis and distance learning.

Conflict of Interest

There is no potential conflict of interest to disclose.

References

• 1)   Sahin  D et al.: Is it possible with smartphone images? Diagn Cytopathol, 2018; 46: 40–46.
• 2)   Lu  MY et al.: Data-efficient and weakly supervised computational pathology on whole-slide images. Nat Biomed Eng, 2021; 5: 555–570.
• 3)   Hernández-Neuta  I et al.: Smartphone-based clinical diagnostics: Towards democratization of evidence-based health care. J Intern Med, 2019; 285: 19–39.
• 4)   Roy  S et al.: Smartphone adapters for digital photomicrography. J Pathol Inform, 2014; 5: 24.
• 5)   Khurana  KK: Telecytology and its evolving role in cytopathology. Diagn Cytopathol, 2012; 40: 498–502.
• 6)   Pal  A et al.: Deep multiple-instance learning for abnormal cell detection in cervical histopathology images. Comput Biol Med, 2021; 138: 104890.
• 7)   Tanaka  Y et al.: Histologic correlation between smartphone and coloposcopic findings in patients with abnormal cervical cytology: Experiences in a tertiary referral hospital. Am J Obstet Gynecol, 2016; 221: 241.e1–241.e6.
• 8)   Reusche  R et al.: A systematic review of smartphone applications for plastic surgery providers: Target audience, uses, and cost. Ann Plast Surg, 2016; 77: 6–12.