Color Photographs Research Articles

Extended distribution of Trillium govanianum Wall. ex D.Don (Melanthiaceae), an endangered species from Arunachal Pradesh, India

Trillium govanianum Wall. ex D.Don is an endangered herbaceous species under the family Melanthiaceae, endemic to the Himalayan region of India at an elevation between 2,400–3,500 m. The species is mostly distributed in temperate and sub-alpine zones of the Himalaya. This paper reports an extended distribution of T. govanianum Wall. ex D.Don as a new addition to the flora of Arunachal Pradesh, India for the first time. In the present study, the species was recorded from the Mandala region, West Kameng District of Arunachal Pradesh. Five locations of the individuals of the species were reported and distribution was mapped along with a detailed description and colour photographs provided for the easy identification of the species.

Begonia tamoiana (Begoniaceae), a new species from Atlantic Forest of Rio de Janeiro State, Brazil

Begonia tamoiana is a new species endemic to Brazilian Atlantic Rainforest, a biome rich in endemism and species diversity, and it is described and illustrated herein. Begonia tamoiana is morphologically similar to B. magdalenensis but differs in having abaxial leaf blades with simple trichomes throughout (vs. 2-branched trichomes throughout) and ovate wings capsule (vs. triangular). A description, ecological comments, illustrations, color photographs, and a provisional IUCN conservation assessment of this species are provided. Begonia tamoiana is preliminarily classified as Vulnerable (VU) due to its very small populations and restricted distribution.

CNN-Based Integrated Framework for Enhanced Diabetic Retinopathy Detection

Background: Diabetic Retinopathy (DR), a significant cause of vision loss globally, is characterized by retinal damage caused by diabetes. Early detection is vital to prevent irreversible blindness, yet challenges remain in accurately identifying DR stages and enhancing blood vessel visibility in fundus images. This paper aims to develop an early detection methodology for DR, addressing the need for early diagnosis and the difficulties in distinguishing DR severity through fundus imaging. The challenges in the early detection of Diabetic Retinopathy (DR) and enhancing blood vessel visibility in fundus images are multifaceted, including issues such as data imbalance, image noise, and complex patterns. By addressing these challenges through advanced ML techniques and image processing methodologies, the proposed methodology in the paper aims to overcome the limitations in early detection and severity assessment of DR, contributing to improved patient outcomes and vision preservation. Methods: This study utilizes Machine Learning (ML) to analyze complex patterns in fundus color images of DR, employing spatial domain filtering to reduce image noise and address data imbalances across DR severity levels through data augmentation. A Convolutional Neural Network (CNN), enhanced with a Gabor filter, is applied for stage-specific DR detection and to pinpoint infected areas. The dataset includes 1000 color fundus images, with a 70:30 split for training and testing, respectively. The adoption of a Gabor filtering technique aims to refine the model’s performance further. Results: The incorporation of a CNN with a Gabor filter has shown outstanding efficacy in detecting DR from fundus color images, achieving a training accuracy of 97.5%, validation accuracy of 96.5%, Cohen kappa score of 89.76%, and testing accuracy of 95.87%. This method effectively illustrates the disease-affected areas in the fundus images provided. Conclusion: Through a comparative analysis of image processing techniques, this research highlights the advantages of using advanced DR analysis for image preprocessing. The proposed CNN-Gabor filter approach demonstrates significant success in identifying diabetic retinopathy in fundus color photographs and accurately delineating the affected regions, contributing valuable insights to the field of medical image processing.

Coloring in the world of others: color use in visual orientalism, 1890–1920

This study investigates the impact of media on color sense: our ability to see different colors and use them to interpret the world. Specifically, we examine the role of color in the cultural construction of the Orient—an ‘imagined geography’ used to justify colonial domination—in two turn-of-the-twentieth-century types of color(ed) photographs: photochromes, where a printer added color, and autochromes, where colors were captured during exposure. While most research on visual Orientalism has focused on content, we use machine learning methods to study the most important formal element of visual Orientalism: color. After using K-means clustering to extract sixteen dominant colors from each photograph in our dataset, we train three different random forest classification algorithms to make a distinction between (A) the two color media (B) photochromes of the Orient and the Occident; and (C) autochromes of the Orient and the Occident. Subsequently, we apply Shapley Additive Explanations, an explainable AI method, to interpret the output of the classifiers. This allows us to examine how specific features (colors) impacted the classifiers’ predictions. While the algorithm can easily separate photochromes from autochromes (0.95) and Oriental from Occidental photochromes (0.93), it struggles with the same task in the autochrome collection (0.68). These findings support three interconnected conclusions: (1) color sense became mediated in the late nineteenth century, (2) in photochromes, the presence and absence of specific colors was a vital aspect of visual Orientalism, (3) the autochrome, where color was derived from light, provided a more objective picture of countries in the near and middle East than the photochrome.

Taxonomy of the genus Diochus Erichson (Coleoptera: Staphylinidae, Staphylininae, Diochini) of Papua New Guinea, Vanuatu and Solomon Islands

This paper presents taxonomic studies of the genus Diochus Erichson, 1839 (Coleoptera: Staphylinidae, Staphylininae, Diochini) from Papua New Guinea, Vanuatu and Solomon Islands. Three new species are described: Diochus elongatus Huang, Janák & Zhou, sp. nov. and D. loebli Huang, Janák & Zhou, sp. nov. from Papua New Guinea, and D. chambrieri Huang, Janák & Zhou, sp. nov. from Papua New Guinea, Vanuatu and Solomon Islands. D. papuanus Cameron, 1939 is redescribed. All species are included in a key to species and are provided with color photos. The new species also with line drawing plates.

Taxonomic review of the genus Hampsonidia Inoue, 1958 (Lepidoptera: Noctuidae: Acronictinae) with describing two new species from China

AbstractThe genus Hampsonidia Inoue, 1958 (objective replacement name for Hampsonia Kozhanchikov, 1950) was revised. Two new species are described from China: H. georgii Kiss sp. n. and H. rocinante Kiss & H.-L. Han sp. n. All species of the genus are illustrated by colour photos of adults and black-and-white genitalia photos, and in two species, both male and female last abdominal segments are illustrated. The male last abdominal segments of Herzinycta albonigra (Herz, 1904) and the female last abdominal segments of Draudtinycta tigniumbra (Draudt, 1937) are illustrated and described for the first time.

Crinipellis siamensis (Marasmiaceae, Agaricales), a new macrofungus from northern Thailand

A new Crinipellis species, described herein as C. siamensis, was collected from northern Thailand during the rainy seasons in 2022 and 2023. This species was identified based on morphological characteristics and molecular data. Crinipellis siamensis is characterized by its crinipelloid basidiomata, grayish orange to brown pileus, elongated to cylindrical basidiospores, and hair-like terminal cells on the pileus surface. Phylogenetic analysis of combined internal transcribed spacer (ITS) and the large subunit (nrLSU) of nuclear ribosomal DNA (rDNA) regions revealed that C. siamensis is well-supported as a distinct monophyletic member within the genus Crinipellis. A full description, color photographs, illustrations, and a phylogenetic tree showing the position of C. siamensis are provided. Morphologically similar and related species to the new species are also discussed.

Additions to the angiosperm flora of Manipur, India

The investigation into angiosperm flora has identified 5 flowering plants as a new record to Manipur state, northeast India: Amorphophallus napalensis (Wall.) Bogner & Mayo and Arisaema nephanthoides (Wall.) Mart. (Araceae), Impateins pulchra Hook. f. & Thomson (Balsaminaceae), Lonicera acuminata Wall. (Caprifoliaceae) and Silene baccifera (L.) Durande (Caryophyllaceae). Notably, the genus Silene is a new addition to the state’s flora. Detailed morphological descriptions and colour photo plates are provided for easy identification.

Integrative taxonomic revision of the Cerurina-Afrocerura genus-pair (Lepidoptera, Noctuoidea, Notodontidae, Cerurinae)

Abstract The Afrotropical Cerurinae (Notodontidae) genera Cerurina Kiriakoff and Afrocerura Kiriakoff are revised. Historically, significant confusion has existed surrounding the identities of members of these 2 closely related genera. Using morphology and genetic data derived from mitochondrial COI barcodes and anchored hybrid enrichment nuclear genomics, we present an integrated taxonomic approach. Upon examining all types, historical material in global collections, and numerous recently collected specimens, we are able to associate available names with the taxa in these genera as well as describe four new species of Afrocerura for which no names were available. Historical misidentifications are also addressed. For nearly all species, both sexes and their genitalia are figured. In total, we recognize one species in Cerurina with 2 subspecies: C. marshalli marshalli Hampson (type species) and C. marshalli cameroona (Bethune-Baker) comb. n., and 8 species in Afrocerura: A. leonensis (Hampson) (type species), A. takanoisp. n., A. bifasciata (Janse), A. inexpectatasp. n., A. tanganyikae Kiriakoff stat. n., A. ethiopicasp. n., A. thomensis (Talbot) and A. smithisp. n. Color photos of C. m. cameroona larvae are presented for the first time and are compared to larvae of C. m. marshalli.

What is concealed under the name Lochmaea crataegi (Coleoptera: Chrysomelidae)?

Until now, Lochmaea crataegi (Forster, 1771) was considered a transpalaearctic species, while the closely related species L. machulkai Roubal, 1926 was known only from the Caucasus and adjacent regions. The present study establishes that L. crataegi is found exclusively in Western, Central, and Southern Europe, while L. machulkai is distributed in Crimea, the central and southern regions of European Russia, the Caucasus, northern Iran, eastern Turkey, and western Middle Asia. A new species, L. sergeevi sp. nov. , is described from Asian Russia, specifically from the Urals to the Far East. The males of L. sergeevi sp. nov. differ from those of L. crataegi and L. machulkai in the aedeagus, which is much wider than that of L. crataegi , with the maximum width closer located to the apex than in L. machulkai (in dorsal view), and significantly less curved than in both compared species (in lateral view), and in the metatibia which is curved near the mid-length, whereas it is curved closer to the base in both of the compared species. The diagnostic characters of the L. crataegi species–group, as well as those of L. crataegi and L. machulkai , are revised. Color photographs of the diagnostic characters and a distribution map of all species are provided.

Association of EPA and DHA with age-related macular degeneration: a cross-sectional study from NHANES.

This cross-sectional study conducted in the general US population investigated the association between dietary intake of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) and the prevalence of AMD. Data from the National Health and Nutrition Examination Survey (NHANES) were utilized, including 4,842 participants aged 40 years and older. Dietary EPA and DHA intake data were collected through two 24-h dietary recall interviews and adjusted for weight. AMD was determined by a standardized grading system based on the presence of key features of AMD in color photographs of the macula. Multivariate logistic regression and restricted cubic spline models evaluated the associations between dietary EPA and DHA intake and AMD. Subgroup analysis and interaction analysis explored the influence of covariates. A total of 4,842 participants were included. In the multivariate-adjusted model 2, the odds ratios (ORs) with 95% confidence intervals (CIs) for AMD were 0.86 (0.75, 0.99) and 0.88 (0.80, 0.97) per unit increase in dietary EPA and DHA intake, respectively. Interaction testing revealed significant effect modification by age, education, and BMI on the EPA-AMD association, indicating these factors significantly impacted this inverse relationship (p-interaction < 0.05). Similarly, age, education, BMI, and cataract surgery history modified the inverse DHA-AMD association (p-interaction < 0.05). Dose-response analyses demonstrated a negative correlation between dietary EPA and DHA intake with AMD prevalence (p-nonlinearity = 0.184 and 0.548, respectively). Our findings suggested that higher dietary EPA and DHA intake could be associated with lower AMD risk in older US adults. Age, education level, BMI, and history of cataract surgery may influence this inverse association.

Targeting the Tie-2 Receptor With Faricimab in Central Serous Chorioretinopathy: A Case Series Motivated by a Genetic Finding

PurposeTo investigate the effects of faricimab, a bispecific antibody targeting VEGF and Ang-2 (thus increasing Tie-2 activity), in patients with CSC based on a recent genetic study that implicated Tie-2 signaling in CSC pathophysiology. DesignA retrospective interventional multicenter case series. MethodsWe included patients with chronic CSC (persistent or recurrent SRF for ≥6 months) who received at least one faricimab 6mg injection between January 1 2022 and April 1 2024. Study sites included Massachusetts Eye and Ear and University of California San Francisco. Patients with evidence of a choroidal neovascular membrane on color photos, optical coherence tomography (OCT) and/or fluorescein angiography were excluded. 16 eyes (15 patients) met the inclusion criteria. The median central macular thickness at each visit from 52 weeks before to 52 weeks after the first faricimab injection was calculated using automated Heidelberg Spectralis ETDRS subfield measurements. ResultsPrior to treatment with faricimab, CSC had been diagnosed a median of 4.1 years (range 0.9–8) earlier and SRF (and intraretinal fluid [IRF] in a subset) had been continuously present for a median of 30 weeks (range 9–257). Decreases in macular thickness were observed in 14/16 eyes after the first faricimab injection and in 14/16 eyes in the full follow-up period compared with prior, 10 of which experienced complete resolution of SRF following the start of the first series of injections at a median of 4 weeks (range 2—25). One eye worsened after the second injection. The median improvement in macular thickness was 40μm [range -3 to 89.5] (P=0.0007). Upon review of OCT images, reductions in macular thickness were consistent with reductions in SRF and/or IRF. Visual acuity improved by 2 lines or more in 6/16 eyes. ConclusionsIn a retrospective case series of patients with chronic CSC and longstanding SRF, we observed improvement in macular thickness after intravitreal faricimab. While the small number of patients and variable natural history of CSC preclude definitive conclusions, a randomized controlled trial seems warranted.

P191 CO-EXISTING HYPERTENSION: AN INDEPENDENT RISK FACTOR FOR DIABETIC RETINOPATHY IN TYPE-2 DIABETES

Background and Objectives: Hypertension has plausible reason to be show the interaction due to its independent effect on retina. Epidemiological studies also provide clue that for hypertension as a possible effect modifier in the risk development of retinopathy in diabetic patients. It is worth exploring whether subgroup in diabetic population with Hypertension have different risk potential of developing retinopathy. The present study aimed to assess the association of hypertension as a risk factor for diabetic retinopathy (DR) in type 2 diabetes mellitus. Methods: A total of 977 type-2 diabetic patients recruited retrospectively in 2008 from Bangladesh Institute of Research and Rehabilitation in Diabetes, Endocrine and Metabolic Disorder (BIRDEM) based on hospital records for determining the incidence of diabetic retinopathy (DR), who were naı̈ve type-2 diabetes during 1993. At the end of 15 year follow-up 495 patients were diagnosed as DR by retinal colour photography. Data on diabetes and lipid profile with serum creatinine, blood pressure and biophysical measures were obtained at baseline, 5, 10 and 15 years follow-up. Binary logistic regression model was used to assess the associations of co-existing hypertension with retinopathy risk adjusting for clinical, biochemical and anthropometric variables. Results: In all three follow-up points, including at baseline, mean systolic and diastolic blood pressure were found to significantly higher among patients who developed DR (P&lt;.001). Binary logistic regression model confirmed coexisting hypertension (OR 3.2; 95% CI 2.2 -4.7) as a significant risk factor of DR adjusting for possible confounders; serum creatinine (OR 21.9; 95% CI 14.0 -35.4) and glycemic control status (OR 17.2; 95% CI 9.4 -31.4). Interaction between hypertension, glycemic control and raised serum creatinine did not appear to be statistically significant (P&gt;0.05). Conclusions: Co-existing hypertension is a significant risk factor of DR irrespective of nephropathy or glycolic control status among Type 2 diabetes mellitus.

Color Fundus Photography and Deep Learning Applications in Alzheimer Disease

ObjectiveTo report the development and performance of 2 distinct deep learning models trained exclusively on retinal color fundus photographs to classify Alzheimer disease (AD). Patients and MethodsTwo independent datasets (UK Biobank and our tertiary academic institution) of good-quality retinal photographs derived from patients with AD and controls were used to build 2 deep learning models, between April 1, 2021, and January 30, 2024. ADVAS is a U-Net–based architecture that uses retinal vessel segmentation. ADRET is a bidirectional encoder representations from transformers style self-supervised learning convolutional neural network pretrained on a large data set of retinal color photographs from UK Biobank. The models’ performance to distinguish AD from non-AD was determined using mean accuracy, sensitivity, specificity, and receiving operating curves. The generated attention heatmaps were analyzed for distinctive features. ResultsThe self-supervised ADRET model had superior accuracy when compared with ADVAS, in both UK Biobank (98.27% vs 77.20%; P<.001) and our institutional testing data sets (98.90% vs 94.17%; P=.04). No major differences were noted between the original and binary vessel segmentation and between both eyes vs single-eye models. Attention heatmaps obtained from patients with AD highlighted regions surrounding small vascular branches as areas of highest relevance to the model decision making. ConclusionA bidirectional encoder representations from transformers style self-supervised convolutional neural network pretrained on a large data set of retinal color photographs alone can screen symptomatic AD with high accuracy, better than U-Net–pretrained models. To be translated in clinical practice, this methodology requires further validation in larger and diverse populations and integrated techniques to harmonize fundus photographs and attenuate the imaging-associated noise.

Seven New Species of Entoloma Subgenus Cubospora (Entolomataceae, Agaricales) from Subtropical Regions of China.

Entoloma is a relatively large genus in Agaricales, with a rich diversity of species and a wide distribution. In this study, seven new species of Entoloma belonging to the subgenus Cubospora have been identified based on morphological and phylogenetic evidence from subtropical regions of China. Morphologically, E. excavatum is characterized by the yellow, depressed, estriate pileus and medium-sized basidiospores; E. lacticolor is recognized by the white and papillate pileus, adnexed lamellae, and presence of clamp connections; E. phlebophyllum is identified by the pink-to-maroon and estriate pileus, and lamellae with lateral veins; E. rufomarginatum differs from other cuboid-spored species by the lamellae edge which is red-brown-underlined; E. subcycneum is characterized by the white pileus and carneogriseum-type cheilocystidia; E. submurrayi is recognized by the pileus margin exceeding the lamellae, 2-layered pileipellis with hyphae of different widths, and the presence of clamp connections; E. tomentosum is identified by the tomentose pileus, heterogeneous lamella edge, and versiform cheilocystidia with brown-yellow contents. Their distinct taxonomic status is confirmed by the positions of the seven new species in both the ITS + LSU and 3-locus (LSU, tef-1α, rpb2) phylogenetic trees. Detailed descriptions, color photos, and a key to related species are presented.

Selective nanosecond laser removal of retinal pigment epithelium for cell therapy

Retinal pigment epithelial (RPE) cells play a crucial role in the health of the retina, and their dysfunction is associated with various ocular diseases. The transplantation of RPE cells has been proposed as a potential treatment for numerous degenerative diseases, including geographic atrophy from macular degeneration. However, current models to induce RPE damage in animal models prior to transplantation involve mechanical scraping, chemical administration, or laser photocoagulation techniques, which can damage the overlying neurosensory retina. This study aims to investigate the feasibility and efficacy of nanosecond duration laser treatment to safely remove large areas of RPE cells without causing damage to the adjacent tissue or affecting the retinal architecture. Twelve pigmented rabbits were treated with a nanosecond laser on each eye at a laser energy ranging from 200 to 800 nJ with a treated area of 5 × 5 mm2. Human induced pluripotent stem cells-differentiated to RPE (hiPSC-RPE) cells labeled with indocyanine green (ICG), an FDA approved dye, were transplanted subretinally into the damaged RPE areas at day 14 post-laser treatment. The RPE atrophy and hiPSC-RPE cell survival was evaluated and monitored over a period of 14 days using color photography, fluorescein angiography (FA), photoacoustic microscopy (PAM), and optical coherence tomography (OCT) imaging. All treated eyes demonstrated focal RPE loss with a success rate of 100%. The injured RPE layers and the transplanted hiPSC-RPE cells were visualized in three dimensions using PAM and OCT. By performing PAM at an optical wavelength of 700 nm, the location of hiPSC-RPE cells were identified and distinguished from the surrounding RPE cells, and the induced PA signal increased up to 18 times. Immunohistochemistry results confirmed the grafted hiPSC-RPE replaced regions of RPE damage. This novel technique has the potential to serve as an animal model of RPE degeneration, to improve models of RPE transplantation, and may help accelerate translation of this therapeutic strategy for clinical use.

Curcuma ungmensis (Zingiberaceae): a new species from Nagaland, North-East India

Curcuma ungmensis M.Sabu, Hareesh and P.Tiatemsu sp. nov. from Mokokchung district of Nagaland state, India is described here. The new species shows affinities towards C. inodora but distinct in many attributes. Detailed description and distribution are provided along with colour photographs of the new species for easy identification.

Origami-based multifunctional sensing platform for sustainable detection of hazardous magnetic materials

Anthropogenic magnetite (AM) nanoparticles have been identified in the human brain and circulatory system, potentially linked to neurodegenerative and cardiovascular diseases. Specifically, AM and other magnetic nanocontaminants from industrial emissions and brake wear are hazardous components of particulate matter. Such contamination enriches urban soils with magnetite and other magnetic nanocontaminants, which can be absorbed by plants like rice and consequently enter the human body indirectly. Developing accurate and robust AM-sensing platforms is crucial, especially in areas where magnetic contamination threatens ecosystems and human health. Innovative materials, such as magnetoactive smart materials, are essential for creating sensors with specific, wireless, and adjustable magnetic properties for efficient detection and monitoring of soil contamination. This study presents an origami-based multifunctional sensing platform for sustainable detection of magnetic environmental contamination. Utilizing paper as its substrate for low-cost AM sensing, the device incorporates two wax/NdFeB magnets, four hydrophilic channels, and a hydrophilic analysis area, enclosed by hydrophobic wax. Through comprehensive analysis techniques including energy-dispersive X-ray spectroscopy, vibrating sample magnetometry, infrared spectroscopy, and photographic color changes, the device exhibited a detection limit below 156 μg. The platform's versatility, affordability, sustainability, and capacity for multi-analysis indicate promising prospects for developing economically equitable, user-friendly, mechanically robust, and flexible magnetic contamination sensing devices. These devices eliminate the need for complex machinery while delivering rapid, accurate, and precise results tailored to diverse environmental needs, thus promoting sustainable and safe societies.

Multimodal imaging diagnosis and analysis of prognostic factors in patients with adult-onset Coats disease.

To describe the multimodal imaging features, treatment, and outcomes of patients diagnosed with adult-onset Coats disease. This retrospective study included patients first diagnosed with Coats disease at ≥18 years of age between September 2017 and September 2021. Some patients received anti-vascular endothelial growth factor (VEGF) therapy (conbercept, 0.5 mg) as the initial treatment, which was combined with laser photocoagulation as needed. All the patients underwent best corrected visual acuity (BCVA) and intraocular pressure examinations, fundus color photography, spontaneous fluorescence tests, fundus fluorescein angiography, optical coherence tomography (OCT), OCT angiography, and other examinations. BCVA alterations and multimodal image findings in the affected eyes following treatment were compared and the prognostic factors were analyzed. The study included 15 patients who were aged 24-72 (57.33±12.61)y at presentation. Systemic hypertension was the most common associated systemic condition, occurring in 13 (86.7%) patients. Baseline BCVA ranged from 2.0 to 5.0 (4.0±1.1), which showed improvement following treatment (4.2±1.0). Multimodal imaging revealed retinal telangiectasis in 13 patients (86.7%), patchy hemorrhage in 5 patients (33.3%), and stage 2B disease (Shield's staging criteria) in 11 patients (73.3%). OCT revealed that the baseline central macular thickness (CMT) ranged from 129 to 964 µm (473.0±230.1 µm), with 13 patients (86.7%) exhibiting a baseline CMT exceeding 250 µm. Furthermore, 8 patients (53.3%) presented with an epiretinal membrane at baseline or during follow-up. Hyper-reflective scars were observed on OCT in five patients (33.3%) with poor visual prognosis. Vision deteriorated in one patient who did not receive treatment. Final vision was stable in three patients who received laser treatment, whereas improvement was observed in one of two patients who received anti-VEGF therapy alone. In addition, 8 of 9 patients (88.9%) who received laser treatment and conbercept exhibited stable or improved BCVA. Multimodal imaging can help diagnose adult-onset Coats disease. Anti-VEGF treatment combined with laser therapy can be an option for improving or maintaining BCVA and resolving macular edema. The final visual outcome depends on macular involvement and the disease stage.

Multi-class classification of pathological myopia based on fundus photography

Pathological myopia (PM) is a severe ocular disease leading to blindness. As a traditional noninvasive diagnostic method, fundus color photography (FCP) is widely used in detecting PM due to its high fidelity and precision. However, manual examination of fundus photographs for PM is time-consuming and prone to high error rates. Existing automated detection technologies have yet to study the detailed classification in diagnosing different stages of PM lesions. In this paper, we proposed an intelligent system which utilized Resnet101 technology to multi-categorically diagnose PM by classifying FCPs with different stages of lesions. The system subdivided different stages of PM into eight subcategories, aiming to enhance the precision and efficiency of the diagnostic process. It achieved an average accuracy rate of 98.86% in detection of PM, with an area under the curve (AUC) of 98.96%. For the eight subcategories of PM, the detection accuracy reached 99.63%, with an AUC of 99.98%. Compared with other widely used multi-class models such as VGG16, Vision Transformer (VIT), EfficientNet, this system demonstrates higher accuracy and AUC. This artificial intelligence system is designed to be easily integrated into existing clinical diagnostic tools, providing an efficient solution for large-scale PM screening.