Pith. sign in

REVIEW 5 major objections 4 minor 33 references

One ordinary photo of a skin lesion can be turned into metric-scale 3D depth, surface normals, and lesion measurements after training on synthetic dermoscopic data.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · deepseek-v4-flash

2026-08-02 06:08 UTC pith:LISCHANJ

load-bearing objection Solid dataset-plus-fine-tuning paper with a real scale-correction result, but the cross-dataset generalization claim doesn't survive contact with Table 3: DDI improves only after adding DDI pseudo-GT to training. the 5 major comments →

arxiv 2607.13010 v2 pith:LISCHANJ submitted 2026-07-14 cs.CV

DermDepth: Toward Monocular Metric Scale 3D Reconstruction Models for Dermatology

H\'ector Carri\'on , Narges Norouzi This is my paper
classification cs.CV
keywords metric-scale 3D reconstructiondermatologymonocular depth estimationsynthetic dataskin lesion measurementsurface normalsfairnessdermoscopy
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper aims to prove that dermatological measurement—tracking lesion size, volume, and surface texture over time—does not require 3D scanners or multiple captures. The authors construct D-Synth, a synthetic dermoscopic dataset with pixel-perfect metric depth and surface normals, and fine-tune only the scale and normal output heads of an existing affine-invariant monocular depth model on it, followed by a small amount of real clinical data. They report that dermoscopic scale error drops from over 16× to under 1.1× on real images, geometric quality is preserved, normal-map detail increases, and lesion width estimates on 14 real test lesions mostly fall within published clinical size ranges. The significance, if correct, is that a single ordinary photograph could support lesion monitoring, ABCDE screening, and wound measurement in settings where 3D hardware is unavailable.

Core claim

The paper claims that monocular metric-scale 3D reconstruction of skin is achievable by combining a new synthetic dataset with pixel-perfect 3D annotations (D-Synth) and parameter-efficient fine-tuning of a foundation model's scale and normal heads. It shows that all four state-of-the-art metric-depth models tested systematically misestimate scale on dermatological images—overestimating dermoscopic depth by 4–16× and underestimating wound depth—while DermDepth brings scale ratios to 0.87× (dermoscopy), 0.91× (wounds), and 1.95× (ruler-area ratio on DDI), and reduces Fitzpatrick skin-tone fairness disparity from 10.90 to 1.02.

What carries the argument

The load-bearing mechanism is D-Synth, a rendered synthetic dermoscopic dataset providing pixel-perfect metric depth, surface normals, and camera intrinsics for 4,000 lesions across 131 melanin levels and 240 morphologies. DermDepth fine-tunes only the scale-prediction MLP and normal-decoder ConvStack of an existing affine-invariant depth foundation model (0.6% of parameters), using a symmetric log-space scale loss and an angular normal loss. Progressive refinement on real datasets, including 33 DDI samples whose metric scale is anchored by visible medical rulers, extends the correction across imaging distances from millimeters to meters.

Load-bearing premise

The main quantitative claims lean on estimating metric scale by finding and segmenting a medical ruler in 47 real skin images; if that ruler-based scale is inaccurate or those images are not representative, the reported correction to under 1.1× and the fairness numbers do not follow.

What would settle it

Take a new set of at least 50 dermoscopic images with ground-truth metric depth measured by a structured-light or plenoptic scanner (not a ruler), run DermDepth, and compute the median predicted-to-true depth ratio; if it falls outside roughly 0.9×–1.1×, the single-image metric-scale claim is falsified on real data.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • If DermDepth is right, dermatologists could record lesion width, area, and volume from a single photograph, enabling longitudinal tracking in primary care and telemedicine without 3D hardware.
  • The reported scale correction (16.1× error to about 1×) implies domain-specific fine-tuning can adapt metric-depth foundation models to dermatology, while the raw baselines are not clinically usable for measurement.
  • Improved surface-normal maps could supply quantitative texture and border-morphology features for ABCDE screening, supplementing 2D image analysis.
  • Fairness results (disparity 10.90 → 1.02) suggest that diverse synthetic skin tones plus a small real-sample fine-tune can reduce skin-tone bias in 3D dermatology models.
  • With only 2.1M trainable parameters, the model is lightweight enough that metric 3D measurement could run on mobile devices in low-resource settings.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The DDI quantitative claims rest on ruler-derived pseudo ground truth; an independent dataset with true 3D scans (e.g., structured light) would be the definitive next validation, and the paper's own limitation statement asks for more clean 3D data.
  • Because D-Synth is rendered only at 12–20 mm dermoscopic distances, extending the synthetic renderer to macroscopic ranges could make sim-to-real transfer self-sufficient and reduce dependence on difficult-to-collect real 3D data.
  • The reported normal-map texture improvement is qualitative; a quantitative descriptor (e.g., normal-map roughness or anisotropy) could be tested against clinical features such as ulceration or papillomatous morphology.
  • The same output-head fine-tuning recipe might transfer to other medical imaging domains that can generate synthetic pixel-perfect depth, such as endoscopy or retinal imaging.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

5 major / 4 minor

Summary. Carrión and Norouzi introduce DermDepth, a fine-tuning approach for metric-scale monocular 3D reconstruction in dermatology. They construct D-Synth, a 4,000-sample synthetic dermoscopic dataset with rendered metric depth and normal maps. Starting from MoGe-2, they freeze the encoder and train a 3-layer MLP scale head and a ConvStack normal head on D-Synth, then progressively fine-tune the scale head on real data: SKINL2 (263 train), WoundsDB (50), and 33 DDI pseudo-GT samples with ruler-derived scale. On held-out test splits, scale ratios improve from 16.10 to 0.87 (SKINL2), 0.62 to 0.91 (WoundsDB), and the DDI ruler-area ratio from 81.0 to 1.95, while SI-δ1 remains high. The paper also reports normal-map visual improvements, a Fitzpatrick skin-tone disparity reduction (10.90 to 1.02), and lesion width/area/volume estimates broadly consistent with published size ranges.

Significance. The paper's strongest contribution is empirical: it documents that four metric-depth foundation models fail badly on dermatological images (4–16x overestimation on dermoscopy, 0.6–0.75x underestimation on wounds), and that fine-tuning only 0.6% of parameters on a synthetic dataset corrects dermoscopic scale error to 1.11x. The release of D-Synth with pixel-perfect depth and normals, and the systematic evaluation across four baselines, are useful resources for the community. The fairness analysis is a positive addition. However, the significance is tempered by the fact that the reported cross-benchmark results are not truly out-of-distribution: the final model trains on train splits of all three evaluation datasets, and the DDI result requires DDI pseudo-GT labels. The synthetic-only benefit for WoundsDB is negative (0.28x vs 0.62x baseline). If the authors provide a true held-out evaluation or clearly frame the method as per-dataset fine-tuning, the contribution would be solid but more modest.

major comments (5)
  1. [Abstract; Table 3, rows 5–6] The generalization claim is stronger than the evidence. The DDI ratio improves to 1.95 only after adding 33 DDI pseudo-GT images to training (Table 3, row 6). With D-Synth plus real depth from SKINL2 and WoundsDB, the DDI ratio remains 18.3x (Table 3, row 5), still far from metric scale. Since the final model was fine-tuned on train splits of all three benchmarks, Table 2 measures per-dataset calibration, not cross-domain generalization. I recommend either (a) reporting a zero-shot result on a dataset not used in training, or (b) explicitly rephrasing the abstract and conclusion to say the method requires per-dataset fine-tuning (including pseudo-GT) rather than 'generalizes across three benchmarks'.
  2. [§2.1, §3.1, Table 2] The DDI validation is partially circular. Pseudo-GT scale is derived by segmenting a medical ruler and computing its area; the same ruler-area ratio ρ is used as the evaluation metric. Training on these labels may teach the scale head to exploit ruler-specific cues (e.g., ruler presence, known area 6.6 cm²) rather than producing a generally metric depth. The paper acknowledges segmentation uncertainty in §4 but provides no sensitivity analysis. I ask for an error analysis linking ruler segmentation accuracy to ρ, and ideally an independent check (e.g., known dimensions of other objects in the scene) to validate the DDI scale.
  3. [Tables 2–4] No confidence intervals or significance tests are reported. The DDI test set has n=14 and WoundsDB n=27; per-Fitzpatrick groups in Table 4 have at most a handful of samples. The disparity reduction (10.90 to 1.02) and scale-ratio improvements could be within noise. Please report bootstrap confidence intervals or per-sample distributions, especially for DDI and the fairness table.
  4. [§3.5, Figure 3; Abstract] The claim of 'increasing texture richness' is supported only by qualitative visual inspection. No quantitative normal-map error or surface-detail metric is reported on real data; the normal head is frozen during real-data training because ground-truth normals are noisy. Provide a quantitative normal evaluation on D-Synth test or another reliable source, or qualify the claim to say the improvement is visual/qualitative.
  5. [§3.6, Table 5] The lesion-width validation against literature ranges is presented as evidence of metric accuracy, but 4 of 14 predictions fall outside the cited ranges and are rationalized post hoc as 'boundary cases.' No uncertainty estimates are given for the predicted widths. This is a preliminary sanity check, not a quantitative validation of metric accuracy. I suggest reporting error bars and a pre-defined tolerance criterion for agreement.
minor comments (4)
  1. [Figure 2 (left)] The 'Scale Error Training Trajectory' plot is not described. Please clarify what the plotted curves/histograms represent (median, distribution, or individual samples) and how the training stages map to the curves.
  2. [§3.2] SI-AbsRel=0.017 is mentioned once but not defined in the metrics list nor reported for other models. Either define and report it consistently or remove it.
  3. [Reference [30]] Reference [30] appears to be a generic segmentation paper and does not directly support the claim about difficulty of collecting 3D dermatology datasets. Consider citing a dataset-collection-specific reference.
  4. [§4, Limitations] The limitations paragraph acknowledges ruler segmentation uncertainty and the small DDI test set, but it should also state explicitly that the DDI result was obtained after training on DDI pseudo-GT data and therefore does not demonstrate cross-dataset generalization.

Circularity Check

0 steps flagged

No material circularity: core sim-to-real scale result is independent; DDI numbers are transparent in-distribution fine-tuning.

full rationale

The paper's principal scale correction claim is not circular. DermDepthS is trained only on D-Synth rendered depth and evaluated on real SKINL2 dermoscopy, so the 16.1x to 1.11x improvement is an actual synthetic-to-real transfer, with SKINL2 ground truth used solely for evaluation at that stage. The final DermDepth numbers for SKINL2, WoundsDB, and DDI are obtained after fine-tuning on train splits of those same datasets, but the paper discloses this explicitly (Table 3, Section 3.3) and evaluates on held-out test splits, which is standard supervised evaluation rather than a prediction forced by construction. The DDI pseudo-ground-truth is derived from ruler segmentation masks taken from the authors' prior FEDD work [9]; this is a genuine data-dependency and a limitation (acknowledged in Section 4), and it means the DDI fairness/lesion measurements inherit any systematic segmentation bias of [9], but it is not an equation-level equivalence between the model's inputs and its outputs. No self-definitional step, no ansatz smuggled through citation, and no imported uniqueness theorem was found. The only caveat is that DDI should be understood as in-distribution fine-tuning validation, not zero-shot cross-dataset generalization; the abstract's 'generalizes across benchmarks' is somewhat optimistic but does not amount to circularity.

Axiom & Free-Parameter Ledger

4 free parameters · 5 axioms · 0 invented entities

DermDepth is an empirical fine-tuning result. The central claim rests on the validity of the training and evaluation data: D-Synth's simulated metric labels, SKINL2/WoundsDB sensor-derived 3D, and especially DDI ruler-based pseudo-GT. No new physical entities are introduced. The main burden is that benchmark results are in-distribution (fine-tuned on splits of the same datasets) and partly fitted via DDI pseudo-labels.

free parameters (4)
  • D-Synth capture configuration (distance 12–20 mm, FoV 75°, 4,000 samples) = 12–20 mm; 75°; 4000
    Chosen to match dermoscopic range (§2.1). The synthetic-only model degrades WoundsDB scale to 0.28×, showing this configuration does not transfer to macroscopic range.
  • DDI pseudo-GT inclusion rule (fully visible ruler), n=47/33 = 47 selected; 33 train
    Hand-applied subsetting (§2.1, §2.3) that defines both DDI training labels and the DDI test set; no selection or segmentation-error analysis.
  • Training hyperparameters (LR 1e-4/5e-5, 1000/2000 steps, batch 4, EMA 0.999)
    Chosen empirically (§2.2); no sensitivity analysis; EMA is stated as essential for stable convergence.
  • Scale/normal loss weighting
    Equations (1)–(2) define the losses but no combination weights are reported (§2.2); the final objective is underspecified.
axioms (5)
  • domain assumption Monocular metric scale is identifiable from image content alone.
    The scale head regresses a scalar from the CLS token (§2.2). If scale is not visually identifiable, no fine-tuning can produce metric depth; SKINL2/WoundsDB results support this empirically but no justification is given.
  • domain assumption Ruler-derived pseudo-GT accurately represents DDI scene scale.
    §2.1/§2.3: known ruler dimensions plus segmentation masks estimate metric scale; used as both training signal and evaluation ground truth.
  • domain assumption SKINL2/WoundsDB sensor noise does not affect metric scale.
    §2.1 asserts this without testing; ToF spatial offset or planar noise could bias scale estimates.
  • domain assumption Synthetic D-Synth renderings transfer to real dermoscopic images.
    Core sim-to-real premise; the paper acknowledges a photorealism gap but does not quantify its effect on scale learning.
  • domain assumption Published clinical size ranges are appropriate reference standards for validating predicted widths.
    §3.6: 10 of 14 predictions fall within cited literature ranges; 4 are reclassified as 'boundary cases' post hoc, weakening the validation.

pith-pipeline@v1.3.0-alltime-deepseek · 8343 in / 18492 out tokens · 159527 ms · 2026-08-02T06:08:21.237669+00:00 · methodology

0 comments
read the original abstract

Dermatological practice routinely involves measuring and tracking lesion size, morphology and texture, as critical components of wound or skin cancer screening, monitoring and diagnosis. To accomplish this task, practitioners often image the skin surface with commonly available off-the-shelf camera sensors. This has led to an overwhelming research focus on 2D methods while these objectives naturally benefit from 3D information. In this paper, we demonstrate that dense monocular 3D reconstructions, metric scale measurements and rich surface normal texture estimates are achievable for both dermoscopic and macroscopic cases without the need for additional hardware or multiple captures. We present DermDepth, the first single-view metric scale 3D model for the dermatological domain and D-Synth, the first synthetic dermoscopic dataset with pixel-perfect 3D information. Our experiments show training DermDepth on D-Synth corrects metric scale error from over 16x to under 1.1x for real dermoscopic data, while preserving geometric quality and increasing texture richness. Fine-tuning on a small amount of real clinical samples generalizes our method across three real-world benchmarks spanning the few mm to hundred cm range, diverse skin-tones, chronic wound cases and produces measurements broadly consistent with disease size reported in medical literature. All code, data and models are available at https://github.com/hectorcarrion/dermdepth.

Figures

Figures reproduced from arXiv: 2607.13010 by H\'ector Carri\'on, Narges Norouzi.

Figure 1
Figure 1. Figure 1: D-Synth Dataset. Rendered skin-lesion synthetics from our D-Synth pipeline with dense pixel-perfect 3D information and texture normal maps. Color, Diameter and Evolution (ABCDE measurements) in dermatology [8]; fea￾tures that Artificial Intelligence (AI) algorithms could facilitate tracking. As such, dermatology is often a measurement problem: clinicians screen, triage, and monitor lesions or wounds by tra… view at source ↗
Figure 2
Figure 2. Figure 2: Scale Correction Training (Left), Predicted Volume (Right). [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Predicted Depth, Scale and Normals. We visualize performance across three datasets at different metric distances. We find DermDepth produces more accu￾rate and consistent scale and normal maps with finer lesion detail. We note DDI was not collected alongside 3D information, thus we estimate scale error from known ruler area. Ground-truth may show dataset-specific sensor noise, as discussed in Sect. 2.1. 3.… view at source ↗

Review history (2 revisions) →

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

33 extracted references · 4 linked inside Pith

  1. [1]

    Delasco photo measuring labels.https://www.delasco.com/products/ photo-measuring-labels, accessed: 2026-02-23

    2026

  2. [2]

    In: Proceedings of the International Conference on Medical Image Com- puting and Computer-Assisted Intervention (MICCAI) ISIC Skin Image Analysis Workshop

    Abhishek,K.,Hamarneh,G.:Lesionelevationpredictionfromskinimagesimproves diagnosis. In: Proceedings of the International Conference on Medical Image Com- puting and Computer-Assisted Intervention (MICCAI) ISIC Skin Image Analysis Workshop. vol. 15274, pp. 45–55 (2025)

    2025

  3. [3]

    Aboud, A.M.A., Nigam, P.K.: Wart (2023),https://www.ncbi.nlm.nih.gov/ books/NBK431047/ 10 Carrión et al

    2023

  4. [4]

    Archives of Dermatology134(12), 1563–70 (1998)

    Argenziano,G.,Fabbrocini,G.,Carli,P.,etal.:Epiluminescencemicroscopyforthe diagnosis of doubtful melanocytic skin lesions. Archives of Dermatology134(12), 1563–70 (1998)

    1998

  5. [5]

    gov/books/NBK482210/

    Baigrie, D., Tanner, L.S.: Dysplastic nevi (2022),https://www.ncbi.nlm.nih. gov/books/NBK482210/

    2022

  6. [6]

    Journal of Clinical Oncology27(36), 6199– 6206 (2009)

    Balch, C.M., Gershenwald, J.E., Soong, S.J., et al.: Final version of 2009 AJCC Melanoma Staging and Classification. Journal of Clinical Oncology27(36), 6199– 6206 (2009)

    2009

  7. [7]

    JDDG Jour- nal der Deutschen Dermatologischen Gesellschaft21(3), 265–277 (2023)

    Barthelmann, S., Butsch, F., Lang, B.M., et al.: Seborrheic keratosis. JDDG Jour- nal der Deutschen Dermatologischen Gesellschaft21(3), 265–277 (2023)

    2023

  8. [8]

    Journal of the American Academy of Dermatology48(5), 672– 678 (2003)

    Blum, A., Rassner, G., Garbe, C.: Modified ABC-point list of dermoscopy: A sim- plified and highly accurate dermoscopic algorithm for the diagnosis of cutaneous melanocytic lesions. Journal of the American Academy of Dermatology48(5), 672– 678 (2003)

    2003

  9. [9]

    arXiv (Cornell University) (2023), https://arxiv.org/abs/2307.11654

    Carrión, H., Norouzi, N.: FEDD – Fair, Efficient, and diverse diffusion-based lesion segmentation and malignancy Classification. arXiv (Cornell University) (2023), https://arxiv.org/abs/2307.11654

    Pith/arXiv arXiv 2023

  10. [10]

    Telemedicine and e-Health25, 1022–1032 (2019)

    Coustasse, A., Sarkar, R., Abodunde, B., et al.: Use of teledermatology to improve dermatological access in rural areas. Telemedicine and e-Health25, 1022–1032 (2019)

    2019

  11. [11]

    Science Advances8(2022)

    Daneshjou, R., Vodrahalli, K., Novoa, R.A., et al.: Disparities in dermatology ai performance on a diverse, curated clinical image set. Science Advances8(2022)

    2022

  12. [12]

    IEEE2019, 3905–3908 (2019)

    De Faria, S.M.M., Henrique, M., Filipe, J.N., et al.: Light Field Image Dataset of Skin Lesions. IEEE2019, 3905–3908 (2019)

    2019

  13. [13]

    Dermatology240(3), 462–467 (2024)

    Dessinioti, C., Plaka, M., Befon, A., et al.: A retrospective study of diameter and breslow thickness in invasive melanomas. Dermatology240(3), 462–467 (2024)

    2024

  14. [14]

    Dermatology Practical & Conceptual11(2), e2021046 (2021)

    Fougelberg, J., Ek, H., Claeson, M., et al.: Surgery for Bowen Disease: Clinico- pathological factors associated with incomplete excision. Dermatology Practical & Conceptual11(2), e2021046 (2021)

    2021

  15. [15]

    nih.gov/books/NBK545285/

    Greco, M.J., Bhutta, B.S.: Seborrheic keratosis (2024),https://www.ncbi.nlm. nih.gov/books/NBK545285/

    2024

  16. [16]

    Archives of Dermatological Research317(1), 485 (2025)

    Hamzelou, S., Aryanian, Z., Maqbool, F., et al.: Assessment of clinical and patho- logical characteristics of inverted follicular keratosis - five years comprehensive review. Archives of Dermatological Research317(1), 485 (2025)

    2025

  17. [17]

    Advances in Dermatology and Allergology5(5), 375–380 (2016)

    İyidal, A.Y., Gül, Ü., Kılıç, A.: Number and size of acquired melanocytic nevi and affecting risk factors in cases admitted to the dermatology clinic. Advances in Dermatology and Allergology5(5), 375–380 (2016)

    2016

  18. [18]

    IEEE Access9, 7894–7907 (2020)

    Juszczyk, J.M., Wijata, A., Czajkowska, J., et al.: Wound 3D geometrical feature estimation using poisson reconstruction. IEEE Access9, 7894–7907 (2020)

    2020

  19. [19]

    In: International Conference on 3D Vision (3DV)

    Keetha, N., Müller, N., Schönberger, J., et al.: MapAnything: Universal feed- forward metric 3D reconstruction. In: International Conference on 3D Vision (3DV). IEEE (2026)

    2026

  20. [20]

    arXiv (Cornell University) (2024),https://arxiv.org/abs/2408.00191

    Kim, A., Saharkhiz, N., Sizikova, E., Lago, M., Sahiner, B., Delfino, J., Badano, A.: S-SYNTH: Knowledge-Based, Synthetic generation of skin images. arXiv (Cornell University) (2024),https://arxiv.org/abs/2408.00191

    Pith/arXiv arXiv 2024

  21. [21]

    arXiv preprint arXiv:2511.10647 (2025),https://arxiv.org/abs/2511.10647

    Lin, H., Chen, S., Liew, J.H., Chen, D.Y., Li, Z., Shi, G., Feng, J., Kang, B.: Depth anything 3: Recovering the visual space from any views. arXiv preprint arXiv:2511.10647 (2025),https://arxiv.org/abs/2511.10647

    Pith/arXiv arXiv 2025

  22. [22]

    Multidimen- sional Systems and Signal Processing33(2), 1–33 (2022) DermDepth 11

    Lourenco, R.M., Tavora, L.M.N., Assuncao, P.A.A., et al.: Enhancement of light field disparity maps by reducing the silhouette effect and plane noise. Multidimen- sional Systems and Signal Processing33(2), 1–33 (2022) DermDepth 11

    2022

  23. [23]

    Muradia, I., Khunger, N., Yadav, A.K.: A clinical, dermoscopic, and histopatho- logical analysis of common acquired melanocytic nevi in skin of color (2022), https://pmc.ncbi.nlm.nih.gov/articles/PMC9586527/

    2022

  24. [24]

    Transactions on Machine Learning Research (2024)

    Oquab, M., Darcet, T., Moutakanni, T., et al.: DINOv2: Learning robust visual features without supervision. Transactions on Machine Learning Research (2024)

    2024

  25. [25]

    JMIR Dermatology 7, e59839 (2024)

    Parekh, P., Oyeleke, R., Vishwanath, T.: The Depth Estimation and Visualization of Dermatological Lesions: Development and Usability Study. JMIR Dermatology 7, e59839 (2024)

    2024

  26. [26]

    Medical Image Analysis75, 102254 (2021)

    Pereira, P.M.M., Thomaz, L.A., Tavora, L.M.N., et al.: Melanoma classification using light-Fields with morlet scattering transform and CNN: Surface depth as a valuable tool to increase detection rate. Medical Image Analysis75, 102254 (2021)

    2021

  27. [27]

    Medical Image Analysis p

    Sinha, A., Kawahara, J., Pakzad, A., et al.: Dermsynth3d: Synthesis of in-the-wild annotated dermatology images. Medical Image Analysis p. 103145 (2024)

    2024

  28. [28]

    Chinese Medical Journal138(2), 185–192 (2024)

    Wang, M., Gao, X., Zhang, L.: Recent global patterns in skin cancer incidence, mortality, and prevalence. Chinese Medical Journal138(2), 185–192 (2024)

    2024

  29. [29]

    Wang, R., Xu, S., Dong, Y., Deng, Y., Xiang, J., Lv, Z., Sun, G., Tong, X., Yang, J.: Moge-2: Accurate monocular geometry with metric scale and sharp details (2025), https://arxiv.org/abs/2507.02546

    Pith/arXiv arXiv 2025

  30. [30]

    Xie, B.: A Comparative Study of Medical Image Segmentation via ResNet-Driven UNet Architecture. IEEE pp. 96–101 (2025)

    2025

  31. [31]

    arXiv preprint arXiv:2510.07316 (2025),https://arxiv.org/abs/2510.07316

    Xu, G., Lin, H., Luo, H., Wang, X., Yao, J., Zhu, L., Pu, Y., Chi, C., Sun, H., Wang, B., et al.: Pixel-perfect depth with semantics-prompted diffusion transform- ers. arXiv preprint arXiv:2510.07316 (2025),https://arxiv.org/abs/2510.07316

    arXiv 2025

  32. [32]

    Cognitive Computation18(1) (2025)

    Zhang, C., Tan, Z., Yang, X., et al.: From 2D images to 3D model: Segmenting and reconstructing acne for robust skin lesion analysis. Cognitive Computation18(1) (2025)

    2025

  33. [33]

    Zito,P.M.,Scharf,R.:Keratoacanthoma(2023),https://www.ncbi.nlm.nih.gov/ books/NBK499931/

    2023