Table Info

Table 1.

Photography-based diagnostic models

Author, year	Task; classes (n)	Feature extractors/Features extracted	Classifier	Accuracy	Specificity (TNR)	Sensitivity (recall)	Precision (PPV)	AUC	F1-score or Jaccard index
Camalan et al. [1], 2021	Classification; suspicious (54) and normal (54) ROIs in photographic images	-	Inception ResNet-v2	86.5%	-	-	-	-	-
Camalan et al. [1], 2021		-	ResNet-101	79.3%	-	-	-	-	-
Figueroa et al. [2], 2022	Classification; suspicious (i.e., OSCC and OPMD) (~ 2,800) and normal (~ 2,800) photographic images	-	GAIN network	84.84%	89.3%	76.6%	-	-	-
Flügge et al. [3], 2023	Classification; OSCC (703) and normal (703) photographic images	-	Swin-transformer DL network	0.98	0.98	0.98	-	-	0.98
Jubair et al. [4], 2022	Classification; suspicious [i.e., OSCC and OPMD (236)] and benign (480) photographic images	-	EfficientNetB0	85%	84.5%	-	-	0.92	-
Jurczyszyn et al. [5], 2020	Classification; OSCC (35) and normal (35) photographic images (1 normal and one of leukoplakia in the same patient)	MaZda software/Textural features, as run length matrix (two), co-occurrence matrix (two), Haar Wavelet transformation (two)	Probabilistic neural network	-	97%	100%	-	-	-
Lim et al. [6], 2021	Classification; no referral (493), refer—cancer/high-risk (636), refer—low-risk (685), and refer—other reasons (641)	-	ResNet-101	-	-	61.70%	61.96%	-	61.68%
Shamim et al. [7], 2019	Classification; benign and precancerous (200) photographic images	-	VGG19	98%	97%	89%	-	-	-
			AlexNet	93%	94%	88%	-	-	-
			GoogLeNet	93%	88%	80%	-	-	-
			ResNet50	90%	96%	84%	-	-	-
			Inceptionv3	93%	88%	83%	-	-	-
			SqueezeNet	93%	96%	85%	-	-	-
	Classification; types of tongue lesions (300) photographic images	-	VGG19	97%	-	-	-	-	-
			AlexNet	83%	-	-	-	-	-
			GoogLeNet	88%	-	-	-	-	-
			ResNet50	97%	-	-	-	-	-
			Inceptionv3	92%	-	-	-	-	-
			SqueezeNet	90%	-	-	-	-	-
Sharma et al. [8], 2022	Classification; OSCC (121), OPMD (102) and normal (106) photographic images	-	VGG19	76%	-	OSCC: 0.43	OSCC: 0.76	OSCC: 0.92	OSCC: 0.45
					-	Normal: 1	Normal: 0.9	Normal: 0.99	Normal: 0.95
					-	OPMD: 0.78	OPMD: 0.7	OPMD: 0.88	OPMD: 0.74
			VGG16	72%	-	-	-	OSCC: 0.94	-
					-	-	-	Normal: 0.96	-
					-	-	-	OPMD: 0.92	-
			MobileNet	72%	-	-	-	OSCC: 0.88	-
					-	-	-	Normal: 0.99	-
					-	-	-	OPMD: 0.80	-
			InceptionV3	68%	-	-	-	OSCC: 0.88	-
					-	-	-	Normal: 0.1	-
					-	-	-	OPMD: 0.88	-
			ResNet50	36%	-	-	-	OSCC: 0.43	-
					-	-	-	Normal: 0.33	-
					-	-	-	OPMD: 0.42	-
Song et al. [9], 2021	Classification; malignant (911), premalignant (1,100), benign (243) and normal (2,417) polarized white light photographic images	-	VGG19	80%	-	79%	83%	-	81%
Song et al. [10], 2023	Classification; suspicious (1,062), normal (978) photographic images	-	SE-ABN	87.7%	88.6%	86.8%	87.5%	-	-
Song et al. [10], 2023			SE-ABN + manually edited attention maps	90.3%	90.8%	89.8%	89.9%	-	-
Tanriver et al. [11], 2021	Segmentation, object detection and classification; carcinoma (162), OPMD (248) and benign (274) photographic images	-	EfficientNet-b4	-	-	85.5%	86.9%	-	85.8%
			Inception-v4	-	-	85.5%	87.7%	-	85.8%
			DenseNet-161	-	-	84.1%	87.9%	-	84.4%
			ResNet-152	-	-	81.2%	82.6%	-	81.1%
			Ensemble	-	-	84.1%	84.9%	-	84.3%
Thomas et al. [12], 2013	Classification; 192 sections of photographic images from 16 patients	GLCM, GLRL and intensity based first order features (eleven selected features)	Backpropagation based ANN	97.92%	-	-	-	-	-
Warin et al. [13], 2021	Object detection and classification; OPMD (350) and normal (350) photographic images	-	DenseNet-121	-	100%	98.75%	99%	0.99	99%
Warin et al. [14], 2022	Object detection and classification; OPMD (315) and OSCC (365) photographic images	-	DenseNet-169	-	OSCC: 99%	OSCC: 99%	OSCC: 98%	OSCC: 1	OSCC: 98%
			DenseNet-169	-	OPMD: 97%	OPMD: 95%	OPMD: 95%	OPMD: 0.98	OPMD: 95%
			ResNet-101	-	OSCC: 94%	OSCC: 92%	OSCC: 96%	OSCC: 0.99	OSCC: 94%
			ResNet-101	-	OPMD: 94%	OPMD: 97%	OPMD: 97%	OPMD: 0.97	OPMD: 97%
Warin et al. [15], 2022	Object detection and classification; OPMD (300) and normal (300) photographic images	-	DenseNet-121	-	90%	100%	91%	0.95	95%
Warin et al. [15], 2022		-	ResNet-50	-	91.67%	98.39%	92%	0.95	95%
Welikala et al. [16], 2020	Object detection and classification; referral (1,054) and non-referral (379) photographic images	-	ResNet-101	-	-	93.88%	67.15%	-	78.30%
Xue et al. [17], 2022	Classification; ruler (440) and non-ruler (2,377) photographic images; first batch (2,817 images/250 patients), second batch (4,331 images/168 patients)	-	ResNetSt	99.6%	99.6%	100%	97.9%	99.6%	98.9%
Xue et al. [17], 2022		-	Vit	99.8%	99.8%	100%	0.98	99.8%	99.5%

ANN: artificial neural network; DL: deep learning; GAIN: guided attention inference; GLCM: gray-level co-occurrence matrix; GLRL: grey level run-length matrix; OPMD: oral potentially malignant disorders; OSCC: oral squamous cell carcinoma; PPV: positive predictive value; ROI: region of interest; TNR: true negative rate; AUC: area under the curv

Declarations

Acknowledgments

During the preparation of this work the authors used ChatGPT (March 14 version) from OpenAI (https://chat.openai.com/chat) in order to specifically review grammar and spelling. After using this tool/service, the authors reviewed and edited the content as needed and took full responsibility for the content of the publication. No large language models/tools/services were used to analyze and draw insights from data as part of the research process.

Author contributions

ALDA: Conceptualization, Investigation, Writing—original draft, Writing—review & editing. CMP: Writing—review & editing. PAV: Writing—review & editing. MAL: Writing—review & editing. ARSS: Conceptualization, Writing—review & editing.

Conflicts of interest

The authors declare that there are no conflicts of interest.

Ethical approval

This study is in accordance with the Declaration of Helsinki and was approved by the Piracicaba Dental Ethical Committee, Registration number: 42235421.9.0000.5418.

Consent to participate

The informed consent to participate in the study was obtained from all participants.

Consent to publication

Informed consent to publication was obtained from relevant participants.

Availability of data and materials

Not applicable.

Funding

This study was financed, in part, by the São Paulo Research Foundation (FAPESP) [#2021/14585-7, #2022/07276-0], Brasil. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Copyright

Publisher’s note

Open Exploration maintains a neutral stance on jurisdictional claims in published institutional affiliations and maps. All opinions expressed in this article are the personal views of the author(s) and do not represent the stance of the editorial team or the publisher.

References

Camalan S, Mahmood H, Binol H, Araújo ALD, Santos-Silva AR, Vargas PA, et al. Convolutional Neural Network-Based Clinical Predictors of Oral Dysplasia: Class Activation Map Analysis of Deep Learning Results. Cancers (Basel). 2021;13:1291. [DOI] [PubMed] [PMC]

Figueroa KC, Song B, Sunny S, Li S, Gurushanth K, Mendonca P, et al. Interpretable deep learning approach for oral cancer classification using guided attention inference network. J Biomed Opt. 2022;27:015001. [DOI] [PubMed] [PMC]

Flügge T, Gaudin R, Sabatakakis A, Tröltzsch D, Heiland M, van Nistelrooij N, et al. Detection of oral squamous cell carcinoma in clinical photographs using a vision transformer. Sci Rep. 2023;13:2296. [DOI] [PubMed] [PMC]

Jubair F, Al-Karadsheh O, Malamos D, Al Mahdi S, Saad Y, Hassona Y. A novel lightweight deep convolutional neural network for early detection of oral cancer. Oral Dis. 2022;28:1123–30. [DOI] [PubMed]

Jurczyszyn K, Gedrange T, Kozakiewicz M. Theoretical Background to Automated Diagnosing of Oral Leukoplakia: A Preliminary Report. J Healthc Eng. 2020;2020:8831161. [DOI] [PubMed] [PMC]

Lim JH, Tan CS, Chan CS, Welikala RA, Remagnino P, Rajendran S, et al. D’OraCa: Deep Learning-Based Classification of Oral Lesions with Mouth Landmark Guidance for Early Detection of Oral Cancer. In: Papież BW, Yaqub M, Jiao J, Namburete AIL, Noble JA, editors. Medical Image Understanding and Analysis. MIUA 2021. Springer, Cham; pp. 408–22.

Shamim MZM, Syed S, Shiblee M, Usman M, Ali S. Automated detection of oral pre-cancerous tongue lesions using deep learning for early diagnosis of oral cavity cancer. arXiv:1909.08987 [Preprint]. 2019 [cited 2025 Jan 17]. Available from: https://arxiv.org/abs/1909.08987

Sharma D, Kudva V, Patil V, Kudva A, Bhat RS. A Convolutional Neural Network Based Deep Learning Algorithm for Identification of Oral Precancerous and Cancerous Lesion and Differentiation from Normal Mucosa: A Retrospective Study. Eng Sci. 2022;18:278–87. [DOI]

Song B, Li S, Sunny S, Gurushanth K, Mendonca P, Mukhia N, et al. Classification of imbalanced oral cancer image data from high-risk population. J Biomed Opt. 2021;26:105001. [DOI] [PubMed] [PMC]

10.

Song B, Zhang C, Sunny S, KC DR, Li S, Gurushanth K, et al. Interpretable and Reliable Oral Cancer Classifier with Attention Mechanism and Expert Knowledge Embedding via Attention Map. Cancers (Basel). 2023;15:1421. [DOI]

11.

Tanriver G, Soluk Tekkesin M, Ergen O. Automated Detection and Classification of Oral Lesions Using Deep Learning to Detect Oral Potentially Malignant Disorders. Cancers (Basel). 2021;13:2766. [DOI] [PubMed] [PMC]

12.

Thomas B, Kumar V, Saini S. Texture analysis based segmentation and classification of oral cancer lesions in color images using ANN. In: 2013 IEEE International Conference on Signal Processing, Computing and Control (ISPCC). 2013 Sep 26-28; Solan, India. IEEE; 2013. pp. 1–5. [DOI]

13.

Warin K, Limprasert W, Suebnukarn S, Jinaporntham S, Jantana P. Automatic classification and detection of oral cancer in photographic images using deep learning algorithms. J Oral Pathol Med. 2021;50:911–8. [DOI] [PubMed]

14.

Warin K, Limprasert W, Suebnukarn S, Jinaporntham S, Jantana P, Vicharueang S. AI-based analysis of oral lesions using novel deep convolutional neural networks for early detection of oral cancer. PLoS One. 2022;17:e0273508. [DOI]

15.

Warin K, Limprasert W, Suebnukarn S, Jinaporntham S, Jantana P. Performance of deep convolutional neural network for classification and detection of oral potentially malignant disorders in photographic images. Int J Oral Maxillofac Surg. 2022;51:699–704. [DOI] [PubMed]

16.

Welikala RA, Remagnino P, Lim JH, Chan CS, Rajendran S, Kallarakkal TG, et al. Automated Detection and Classification of Oral Lesions Using Deep Learning for Early Detection of Oral Cancer. IEEE Access. 2020;8:132677–93. [DOI]

17.

Xue Z, Yu K, Pearlman PC, Pal A, Chen TC, Hua CH, et al. Automatic Detection of Oral Lesion Measurement Ruler Toward Computer-Aided Image-Based Oral Cancer Screening. Annu Int Conf IEEE Eng Med Biol Soc. 2022;2022:3218–21. [DOI] [PubMed] [PMC]

18.

Fu Q, Chen Y, Li Z, Jing Q, Hu C, Liu H, et al. A deep learning algorithm for detection of oral cavity squamous cell carcinoma from photographic images: A retrospective study. EClinicalMedicine. 2020;27:100558. [DOI] [PubMed] [PMC]

19.

González JD, Quintero-Rojas J. Use of Convolutional Neural Networks in Smartphones for the Identification of Oral Diseases Using a Small Dataset. Rev Fac Ing. 2021;30:e11846. [DOI]

20.

Lin H, Chen H, Weng L, Shao J, Lin J. Automatic detection of oral cancer in smartphone-based images using deep learning for early diagnosis. J Biomed Opt. 2021;26:086007. [DOI] [PubMed] [PMC]

21.

Liyanage V, Tao M, Park JS, Wang KN, Azimi S. Malignant and non-malignant oral lesions classification and diagnosis with deep neural networks. J Dent. 2023;137:104657. [DOI] [PubMed]

22.

Song B, Sunny S, Uthoff RD, Patrick S, Suresh A, Kolur T, et al. Automatic classification of dual-modalilty, smartphone-based oral dysplasia and malignancy images using deep learning. Biomed Opt Express. 2018;9:5318–29. [DOI] [PubMed] [PMC]

23.

Song B, Sunny S, Li S, Gurushanth K, Mendonca P, Mukhia N, et al. Bayesian deep learning for reliable oral cancer image classification. Biomed Opt Express. 2021;12:6422–30. [DOI] [PubMed] [PMC]

24.

Song B, Sunny S, Li S, Gurushanth K, Mendonca P, Mukhia N, et al. Mobile-based oral cancer classification for point-of-care screening. J Biomed Opt. 2021;26:065003. [DOI] [PubMed] [PMC]

25.

Talwar V, Singh P, Mukhia N, Shetty A, Birur P, Desai KM, et al. AI-Assisted Screening of Oral Potentially Malignant Disorders Using Smartphone-Based Photographic Images. Cancers (Basel). 2023;15:4120. [DOI] [PubMed] [PMC]

26.

Uthoff RD, Song B, Sunny S, Patrick S, Suresh A, Kolur T, et al. Point-of-care, smartphone-based, dual-modality, dual-view, oral cancer screening device with neural network classification for low-resource communities. PLoS One. 2018;13:e0207493. [DOI] [PubMed] [PMC]

27.

Y D, Ramalingam K, Ramani P, Mohan Deepak R. Machine Learning in the Detection of Oral Lesions With Clinical Intraoral Images. Cureus. 2023;15:e44018. [DOI] [PubMed] [PMC]

28.

Alhazmi A, Alhazmi Y, Makrami A, Masmali A, Salawi N, Masmali K, et al. Application of artificial intelligence and machine learning for prediction of oral cancer risk. J Oral Pathol Med. 2021;50:444–50. [DOI] [PubMed]

29.

Bashir RMS, Shephard AJ, Mahmood H, Azarmehr N, Raza SEA, Khurram SA, et al. A digital score of peri-epithelial lymphocytic activity predicts malignant transformation in oral epithelial dysplasia. J Pathol. 2023;260:431–42. [DOI] [PubMed] [PMC]

30.

Rosma MD, Sameem AK, Basir A, Mazlipah IS, Norzaidi MD. The use of artificial intelligence to identify people at risk of oral cancer: empirical evidence in Malaysian university. Int J Sci Res Educ. 2010;3:10–20.

31.

Ferrer-Sánchez A, Bagan J, Vila-Francés J, Magdalena-Benedito R, Bagan-Debon L. Prediction of the risk of cancer and the grade of dysplasia in leukoplakia lesions using deep learning. Oral Oncol. 2022;132:105967. [DOI] [PubMed]

32.

Ingham J, Smith CI, Ellis BG, Whitley CA, Triantafyllou A, Gunning PJ, et al. Prediction of malignant transformation in oral epithelial dysplasia using machine learning. IOP SciNotes. 2022;3:034001. [DOI]

33.

Liu Y, Li Y, Fu Y, Liu T, Liu X, Zhang X, et al. Quantitative prediction of oral cancer risk in patients with oral leukoplakia. Oncotarget. 2017;8:46057–64. [DOI] [PubMed] [PMC]

34.

Mahmood H, Shephard A, Hankinson P, Bradburn M, Araujo ALD, Santos-Silva AR, et al. Development and validation of a multivariable model for prediction of malignant transformation and recurrence of oral epithelial dysplasia. Br J Cancer. 2023;129:1599–607. [DOI]

35.

Shephard AJ, Bashir RMS, Mahmood H, Jahanifar M, Minhas F, Raza SEA, et al. A fully automated and explainable algorithm for predicting malignant transformation in oral epithelial dysplasia. NPJ Precis Oncol. 2024;8:137. [DOI]

36.

Shimpi N, Glurich I, Rostami R, Hegde H, Olson B, Acharya A. Development and Validation of a Non-Invasive, Chairside Oral Cavity Cancer Risk Assessment Prototype Using Machine Learning Approach. J Pers Med. 2022;12:614. [DOI] [PubMed] [PMC]

37.

Wang X, Yang J, Wei C, Zhou G, Wu L, Gao Q, et al. A personalized computational model predicts cancer risk level of oral potentially malignant disorders and its web application for promotion of non‐invasive screening. J Oral Pathol Med. 2019;49:417–26. [DOI]

38.

Wu MP, Hsu G, Varvares MA, Crowson MG. Predicting Progression of Oral Lesions to Malignancy Using Machine Learning. Laryngoscope. 2023;133:1156–62. [DOI] [PubMed]

39.

Zhang X, Gleber-Netto FO, Wang S, Martins-Chaves RR, Gomez RS, Vigneswaran N, et al. Deep learning-based pathology image analysis predicts cancer progression risk in patients with oral leukoplakia. Cancer Med. 2023;12:7508–18. [DOI] [PubMed] [PMC]

40.

Cai X, Li L, Yu F, Guo R, Zhou X, Zhang F, et al. Development of a Pathomics-Based Model for the Prediction of Malignant Transformation in Oral Leukoplakia. Lab Invest. 2023;103:100173. [DOI]

41.

Araújo ALD, Sperandio M, Calabrese G, Faria SS, Cardenas DAC, Martins MD, et al. Artificial intelligence in healthcare applications targeting cancer diagnosis—part I: data structure, preprocessing and data organization. Oral Surg Oral Med Oral Pathol Oral Radiol. 2025;[Epub ahead of print]. [DOI]

42.

Zhou B, Khosla A, Lapedriza A, Oliva A, Torralba A. Learning Deep Features for Discriminative Localization. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2016 Jun 27-30; Las Vegas, NV, USA. IEEE; 2016. pp. 2921–9. [DOI]

43.

Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D. Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization. In: 2017 IEEE International Conference on Computer Vision (ICCV). 2017 Oct 22-29; Venice, Italy. IEEE; 2017. pp. 618–26. [DOI]

44.

Kapishnikov A, Bolukbasi T, Viégas F, Terry M. XRAI: Better Attributions Through Regions. arXiv:1906.02825 [Preprint]. 2019 [cited 2025 Jan 17]. Available from: https://arxiv.org/abs/1906.02825

45.

Abnar S, Zuidema W. Quantifying Attention Flow in Transformers. arXiv:2005.00928 [Preprint]. 2020 [cited 2025 Jan 17]. Available from: https://arxiv.org/abs/2005.00928

46.

Ribeiro MT, Singh S, Guestrin C. “Why Should I Trust You?”: Explaining the Predictions of Any Classifier. In: Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Demonstrations. San Diego, California: Association for Computational Linguistics; 2016. pp. 97–101. [DOI]

47.

Araújo ALD, de Souza ESC, Faustino ISP, Saldivia-Siracusa C, Brito-Sarracino T, Lopes MA, et al. Clinicians’ perception of oral potentially malignant disorders: a pitfall for image annotation in supervised learning. Oral Surg Oral Med Oral Pathol Oral Radiol. 2023;136:315–21. [DOI]

48.

Casaglia A, DE Dominicis P, Arcuri L, Gargari M, Ottria L. Dental photography today. Part 1: basic concepts. Oral Implantol (Rome). 2016;8:122–9. [DOI] [PubMed] [PMC]

49.

Hunt B, Ruiz A, Pogue B. Smartphone-based imaging systems for medical applications: a critical review. J Biomed Opt. 2021;26:040902. [DOI] [PubMed] [PMC]

50.

Kauark-Fontes E, Araújo ALD, Andrade DO, Faria KM, Prado-Ribeiro AC, Laheij A, et al. Machine learning prediction model for oral mucositis risk in head and neck radiotherapy: a preliminary study. Support Care Cancer. 2025;33:96. [DOI] [PubMed]

51.

Li J, Li W, Sisk A, Ye H, Wallace WD, Speier W, et al. A multi-resolution model for histopathology image classification and localization with multiple instance learning. Comput Biol Med. 2021;131:104253. [DOI]

52.

Collins GS, Reitsma JB, Altman DG, Moons KGM. Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD): The TRIPOD Statement. Circulation. 2015;131:211–9. [DOI]

53.

Collins GS, Moons KGM, Dhiman P, Riley RD, Beam AL, Van Calster B, et al. TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods. BMJ. 2024;385:e078378. [DOI] [PubMed] [PMC]

54.

Araújo ALD, Sperandio M, Calabrese G, Faria SS, Cardenas DAC, Martins MD, et al. Artificial intelligence in healthcare applications targeting cancer diagnosis—part II: interpreting the model outputs and spotlighting the performance metrics. Oral Surg Oral Med Oral Pathol Oral Radiol. 2025;[Epub ahead of print]. [DOI]

55.

Vujovic ŽÐ. Classification Model Evaluation Metrics. J Adv Comput Sci Appl. 2021;12:599–606. [DOI]