The first patient had noticed a palatal mass approximately four years before presentation. It had enlarged painlessly during the preceding six months and had never been treated. Clinical examination found a firm, poorly defined, non-tender mass about 2.5 by 2.5 by 2 cm beneath intact, smooth mucosa with mild redness but no visible ulceration. Non-contrast CT demonstrated an irregular palatal mass. During surgery on October 8, 2020, the lesion was excised and measured 2.0 by 2.0 by 1.5 cm. The operative specimen was submitted for pathological examination, and no major surgical complication was documented in the available records.

Microscopic analysis showed a tumor with unusually varied architecture and multiple cell populations. Glandular cells, mucus within gland-like spaces, and myxoid tissue were mixed with sheets of mucoid and epidermoid cells and solid epithelial nests. Some larger nests had central necrosis and visible mitotic figures. These findings supported the presence of mucoepidermoid carcinoma, while a solid growth pattern and myoepithelial features supported an adenoid cystic carcinoma component. The final diagnosis was palatal carcinoma ex pleomorphic adenoma containing both components. Quantitative mitotic count, reliable margin status, lymphovascular invasion, depth of invasion, overall histological grade and complete pathological stage were not preserved in the archived records.

The second patient was 53 years old and had carried a right parotid-region mass for more than two years before it began enlarging painlessly over six months. The mass measured approximately 5.0 by 5.0 by 3.0 cm on clinical examination and produced visible facial asymmetry with mild weakness at the right corner of the mouth. No enlarged lymph nodes were palpable, and parotid duct secretion was normal. CT showed a well-defined soft-tissue mass in the right parotid gland. A right superficial parotidectomy with tumor excision was performed on March 8, 2019; intraoperatively, the mass measured 4.0 by 4.0 by 3.0 cm and was adherent to facial nerve branches.

Pathology in the parotid specimen revealed myxochondroid stroma, glandular epithelium, myoepithelial cells, mucous and epidermoid cells, mucin pools, cribriform architecture and solid nests with prominent “comedone-like” central necrosis. The authors classified the lesion as carcinoma ex pleomorphic adenoma with mucoepidermoid and salivary duct carcinoma components. Its Ki-67 proliferation index was approximately 15%. As in the palatal case, the records did not reliably establish histological margins, perineural or lymphovascular invasion, depth of invasion, grade or complete pathological stage. Neither patient underwent neck dissection, so pathological nodal status and extranodal extension could not be evaluated.

Both cases illustrate a central limitation of preoperative evaluation. Imaging can describe a salivary gland mass, but it cannot reliably expose every malignant region inside a heterogeneous tumor. The palatal lesion looked irregular on CT, whereas the parotid lesion appeared clearly demarcated, yet neither appearance established its exact pathology. The authors note that fine-needle aspiration and limited biopsy may also miss one component by sampling only part of the lesion. Their differential diagnoses included pleomorphic adenoma and several other benign or malignant salivary gland tumors, but the final classification depended on identifying benign pleomorphic adenoma tissue alongside two separate malignant morphologies. The diagnoses were therefore retrospective, based on examination of the surgically removed tumors rather than clinical or radiological appearances alone.

Immunohistochemistry and special staining supplied supportive evidence but were interpreted alongside the tissue architecture rather than as stand-alone tests. In the palatal tumor, staining for CK34βE12, CK5/6, p63, GFAP and smooth muscle actin was positive, while CD10 and desmin were negative. In the parotid tumor, S-100, CK7, EMA, p63, calponin, estrogen receptor and p53 were positive; CD117 and progesterone receptor were negative, and mucicarmine highlighted mucin production. The authors regarded mucoepidermoid differentiation as primarily morphologic, supported by the mucinous and epidermoid cells, mucin stains and epithelial markers. They similarly based the adenoid cystic and salivary duct components primarily on characteristic histomorphology, using the available stains as corroboration.

Fluorescence in situ hybridization supplied molecular support for the shared pleomorphic adenoma lineage. Break-apart rearrangements of the PLAG1 gene locus were detected in 86% of tumor cells in the palatal case and 85% in the parotid case. The authors interpreted PLAG1 rearrangement as evidence supporting carcinoma ex pleomorphic adenoma rather than as a specific marker for mucoepidermoid carcinoma. This distinction matters because the tumors contained multiple histologies. The combined evidence—clinical history, imaging, broad histologic sampling, immunohistochemistry, special stains and PLAG1 FISH—was used to separate the cases from pure mucoepidermoid carcinoma, adenoid cystic carcinoma, salivary duct carcinoma and other salivary gland malignancies.

Both tumors were grossly excised, and neither patient received postoperative radiotherapy, chemotherapy or another systemic antitumor treatment. The reasons for omitting neck dissection and adjuvant therapy were not documented and were not retrospectively inferred. During available follow-up, no local recurrence or distant metastasis was identified: the palatal patient was followed for three years and the parotid patient for two years. These outcomes describe only the documented observation periods and do not establish a general treatment effect. Detailed imaging, laboratory results and structured patient-reported outcomes were not systematically preserved, while the two cases cannot show whether similar outcomes would follow other patients. The report’s value lies instead in exposing a rare diagnostic pattern and identifying the information lost when pathological records are incomplete.

Recognizing dual malignant components has practical implications even though this report does not test a treatment strategy. Adenoid cystic carcinoma has a recognized tendency toward perineural invasion and distant metastasis, whereas salivary duct carcinoma may display aggressive local behavior and a comparatively high rate of regional lymph-node metastasis. Identifying all components, together with margins, depth of invasion, perineural and lymphovascular invasion and nodal status, may therefore inform risk assessment and multidisciplinary planning. The authors emphasize extensive sampling of resection specimens and careful review of morphologically distinct areas so that one component does not obscure another. Their literature review indicates that most carcinoma ex pleomorphic adenomas contain one malignant component, making the paired histologies reported here exceedingly uncommon. The two cases show that a seemingly routine salivary gland mass can conceal several cancers at once, and that accurate diagnosis can require the full diagnostic chain rather than imaging or a single laboratory marker alone.

Subject of Research: Dual malignant components in carcinoma ex pleomorphic adenoma

Article Title: Carcinoma ex pleomorphic adenoma containing two cancerous components: two case reports

Article References: Mou, Q., Li, D., Qin, D., Li, K., Nie, M., Liu, Y., & Liu, X. (2026). Carcinoma ex pleomorphic adenoma containing two cancerous components: two case reports. Frontiers in Oncology, 16, Article 1981963. https://doi.org/10.3389/fonc.2026.1981963

Image Credits: AI Generated

DOI: 10.3389/fonc.2026.1981963

Keywords: carcinoma ex pleomorphic adenoma, salivary gland cancer, pleomorphic adenoma, mucoepidermoid carcinoma, adenoid cystic carcinoma, salivary duct carcinoma, PLAG1 rearrangement, histopathology, case reports, Carcinoma, pleomorphic, adenoma

The clinical problem at the heart of the study is deceptively narrow but surprisingly stubborn. The parotid gland, the largest of the salivary glands, can develop two benign tumors that together account for the vast majority of non-cancerous parotid masses: pleomorphic adenoma and Warthin’s tumor. Both matter because their management differs, yet on ultrasound their appearances overlap enough that even experienced ultrasonographers can hesitate. Most AI tools pitched at this task have so far lived comfortably inside retrospective datasets, validated on images the algorithm had effectively already seen in development, and never tested in the messier environment where a real radiologist either accepts or ignores the machine’s suggestion. The Korean team, led from the Department of Biomedical Informatics at Korea University College of Medicine together with clinicians at Korea University Ansan and Anam Hospitals, set out to break that pattern.

Architecturally, the system is a hybrid pipeline built around a division of labor. The first stage uses a YOLO-family object detection model, which scans the ultrasound image to locate the suspicious lesion and delineate the region of interest. The second stage is a classification model, based on the ConvNeXt convolutional network architecture, that examines the localized region and renders a verdict: pleomorphic adenoma or Warthin’s tumor. Crucially, the pipeline is confidence-aware. When the detection model’s certainty about the region it has found falls below a pre-specified threshold, set at 0.80 in the final deployment configuration, the system falls back on the classification model to interpret the whole image instead. That fallback mechanism means the AI never has to commit to a localization it does not trust, a small piece of engineering humility with outsized consequences for performance.

The development data came from 949 ultrasound images collected at Korea University Ansan Hospital, with a completely separate external set of 427 images from Korea University Anam Hospital reserved for testing. Rather than assuming their architecture choices were optimal, the researchers explored 35 different combinations of detection and classification models to understand how the two components complement each other. The exploration revealed something subtle: the hybrid pipeline outperformed both of its standalone components in 22 of the 35 pairings on external data, and the benefit was most pronounced when the detection and classification models performed at a similar level. The best configuration achieved a macro-F1 score of 0.838, a balanced measure of precision and recall. When one component was dramatically stronger than the other, the hybrid’s advantage shrank, a finding that carries practical lessons for anyone assembling multi-stage medical AI systems.

The standalone results themselves underscore the diagnostic difficulty. On the external test set, the detection-first pipeline achieved an area under the receiver operating characteristic curve, or AUROC, of 0.811, while the classification model acting alone reached 0.919. AUROC is a standard gauge of a model’s ability to distinguish the two tumor types across all possible decision thresholds, with 1.0 representing perfect discrimination. But the authors argued that neither number alone answers the question that actually matters in a clinic: does the AI make radiologists better? That question demands a different kind of study entirely, one in which human observers read cases first unaided and then with the algorithm whispering over their shoulder, ideally without knowing which is which.

That is precisely what the two-phase observer-performance study delivered. Three board-certified radiologists read 107 external patient cases, three reads per case for a total of 321 interpretations, first on their own and then with AI assistance available through an interface the radiologists could edit. The statistical machinery, a generalized linear mixed model, was chosen because it can properly handle the fact that each radiologist read each case and the repeated measures are entangled. The verdict: pooled diagnostic accuracy climbed from 76.3 percent to 81.6 percent when AI was in the loop, an odds ratio of 1.54 with a 95 percent confidence interval of 1.12 to 2.13 and a p-value of 0.008, meaning the improvement is unlikely to be statistical noise. Just as strikingly, the three readers converged toward each other, with Fleiss’ kappa, a measure of inter-reader agreement that ranges from 0 to 1, rising from 0.475 to 0.611. In plain terms, the AI did not just make the radiologists more accurate; it made them more consistent with one another.

Perhaps the most revealing results concern how the human readers actually behaved, because selective trust is the difference between collaboration and automation bias. The radiologists were not rubber stamps. They followed the AI’s correct predictions 76.2 percent of the time but adopted its incorrect predictions only 51.1 percent of the time, a statistically significant gap that suggests genuine discrimination rather than blind deference. The benefits also concentrated where they were most needed: on harder cases, the gain from AI assistance was larger, which is exactly the profile one wants from a decision-support tool. The system’s explainability layer, using techniques such as Grad-CAM heatmaps to show which image regions drove the classification and D-RISE explanations for the detection model, gave readers visual evidence they could weigh against their own reading of the scan.

There is also an honest asymmetry in the data that the authors did not hide. Sensitivity for Warthin’s tumor jumped by 18.1 percentage points under AI assistance, while sensitivity for pleomorphic adenoma slipped by 9.3 percentage points. The pattern tracks the AI’s own prediction profile: the machine was better at spotting Warthin’s tumors, and readers absorbed that strength along with that weakness. It is a candid illustration that AI assistance reshapes the whole diagnostic landscape rather than uniformly lifting it, and it flags a priority for future iterations of the system, which should aim to rebalance per-class performance before wider deployment.

Why does this particular trial matter beyond the parotid gland? Because prospective, reader-based evaluations of medical AI remain rare, even as algorithms multiply in the radiology literature. A model can post a dazzling AUROC on a curated dataset and still fail to change a single clinical decision. By pre-specifying a single deployment configuration, testing on external cases from a different hospital, and measuring outcomes that track actual human decisions, this study models the kind of evidence regulators, clinicians, and patients should be demanding. The authors were careful to frame the result as support for the system’s potential as a decision-support tool that operates alongside the interpreting physician, not as a replacement, and to call for further evaluation in real-world clinical workflows where scanner variation, workflow pressure, and patient diversity all conspire to degrade performance.

The trajectory from here is familiar but demanding. The system was built and validated within a single national health system, with consent waived for the retrospective components and institutional review board approval at both hospitals, and the deployment threshold of 0.80 was fixed in advance rather than tuned on the test set. Broader multi-center validation, monitoring for performance drift as imaging hardware and patient populations change, and studies of how radiologists interact with editable AI interfaces over months rather than one sitting are the logical next steps. But the core finding stands on its own: when an AI system is designed to know the limits of its confidence and to keep the radiologist firmly in command, the combination reads salivary gland ultrasound better than either the machine or the human alone. That may be the most viral idea in medical AI right now, not that algorithms will replace doctors, but that carefully engineered humility in software can make human judgment measurably, and provably, sharper.

Subject of Research: Human–AI collaboration for AI-assisted differentiation of benign parotid tumors on ultrasound

Article Title: Human–AI interaction in parotid ultrasound diagnosis: a prospective two-phase observer-performance study

Article References: Kim, J., Lee, S., Hwang, H., Ha, S., Kim, Y., Jeong, I., Kim, M.-S., Kang, B., Kim, J. H., Hong, S. J., Lee, Y. H., Kim, Y. S., Baek, S.-K., You, S.-H., Kim, H., Park, H. S., Yoo, Y., Kwack, T. J., Kwon, S. Y., & Lee, H. (2026). Human–AI interaction in parotid ultrasound diagnosis: a prospective two-phase observer-performance study. BMC Medicine. https://doi.org/10.1186/s12916-026-05301-0

Image Credits: AI Generated

DOI: 10.1186/s12916-026-05301-0

Keywords: artificial intelligence, parotid ultrasound, pleomorphic adenoma, Warthin's tumor, deep learning, radiology, observer-performance study, decision support, confidence-aware AI, ConvNeXt, YOLO, BMC Medicine

The model, named GobletNet by its creators, was designed around a principle known as multi-task learning. Rather than building one network to find tumors and a separate network to classify them, the researchers constructed a single end-to-end system with a shared encoder, the portion of the network that converts raw pixel data into increasingly abstract visual features. Because segmentation and classification draw on overlapping visual information, such as tumor boundaries, echotexture, and internal echoes, sharing those features allows each task to reinforce the other. The network performs two jobs simultaneously: it traces the exact outline of the tumor within the ultrasound image, and it assigns the lesion to one of four categories, namely malignant tumors, pleomorphic adenoma, Warthin tumor, or other benign lesions. That four-way distinction matters clinically, because each category points toward a different management strategy, from watchful waiting to nerve-sparing excision to more aggressive resection.

The evidence base behind the model is unusually robust for this field. Data came from five medical centers, with 1,029 patients forming the internal training and testing dataset and two independent external test sets of 229 and 486 patients reserved to evaluate generalizability. External validation is the gold standard for medical artificial intelligence, because models often fail when moved to new hospitals with different scanners, sonographers, and patient populations. GobletNet passed that test with room to spare. On the segmentation task, it achieved Dice similarity coefficients of 0.956, 0.959, and 0.965 across the internal and two external datasets. The Dice coefficient measures the overlap between the model’s predicted tumor outline and the ground truth, with 1.0 representing perfect agreement, so values above 0.95 indicate that the algorithm traces these lesions nearly as cleanly as a human expert annotator.

Classification performance was equally striking. The model reached accuracies of 0.883 on the internal dataset and 0.841 and 0.843 on the two external sets. More telling is the area under the receiver operating characteristic curve, or AUC, which captures how well the model separates each class from the others across all possible decision thresholds. Using a one-versus-rest macro-averaged approach, GobletNet posted AUCs of 0.980, 0.961, and 0.965 on the internal and external datasets respectively. In practical terms, an AUC above 0.95 is generally considered excellent discrimination, and the fact that the number barely dropped when the model encountered patients it had never seen, from hospitals it had never visited, is precisely the kind of generalizability that most medical AI studies struggle to demonstrate. The model also outperformed three single-task comparison models, supporting the authors’ argument that learning segmentation and classification together produces a stronger system than either task alone.

Subgroup analyses added another layer of reassurance. The researchers stratified performance by patient age, gender, and tumor diameter, three variables that could plausibly shift how tumors appear on ultrasound and how well any classifier holds up. Across all of these strata, the model maintained AUCs above 0.950, indicating that its diagnostic power does not quietly collapse for small tumors, for elderly patients, or for any particular demographic group. This kind of stability matters because a model that performs well only on average, while failing systematically in certain subgroups, could introduce dangerous blind spots into clinical practice. The consistency observed here suggests that the features GobletNet learned are genuinely tied to tumor biology and appearance rather than to confounding characteristics of the dataset.

Perhaps the most clinically consequential part of the study is the reader study, in which radiologists with different levels of experience interpreted the ultrasound images with and without the model’s assistance. Unaided, the radiologists’ diagnostic accuracy ranged from 0.405 to 0.560, a sobering figure that reflects how genuinely difficult parotid tumor classification is, even for trained specialists looking at grayscale images. With GobletNet’s assistance, accuracy climbed to between 0.610 and 0.765 across the group. The improvement was significant, and notably it lifted performance across the experience spectrum, suggesting that the model functions less like a replacement for expertise and more like a common diagnostic floor that raises everyone’s baseline. In the same reader study setting, GobletNet itself achieved a one-versus-rest macro-averaged AUC of 0.962, higher than any of the participating radiologists.

The technical achievement here rests on how the network was built and what it was asked to learn. Ultrasound images are notoriously noisy, with speckle artifacts, variable gain settings, and operator-dependent acquisition, all of which complicate automated analysis. Convolutional neural networks, the workhorse architecture behind modern medical image analysis, learn hierarchical features automatically: early layers detect edges and textures, while deeper layers assemble those primitives into recognizable structures. By training a shared encoder on both a segmentation loss, which penalizes inaccurate tumor boundaries measured through metrics like the Dice coefficient and Intersection over Union, and a classification loss, which penalizes incorrect diagnostic labels, GobletNet is forced to learn representations that are simultaneously spatially precise and diagnostically informative. The authors evaluated the segmentation output with Intersection over Union as well as Dice, and classification with accuracy, macro AUC, F1-score, and Cohen’s kappa, a statistic that corrects for agreement expected by chance and thus provides a stricter measure of true diagnostic skill.

The clinical context explains why this matters. Preoperative differentiation of parotid tumors currently relies on a patchwork of tools. Ultrasound is typically the first-line examination because it is inexpensive, radiation-free, and widely available, but its diagnostic specificity for tumor type is limited. Computed tomography and magnetic resonance imaging offer more detail at greater cost and lower accessibility. Fine needle aspiration can provide cytology but is invasive, occasionally inconclusive, and technically challenging in the parotid region where the facial nerve lies close by. A reliable algorithm that upgrades the diagnostic yield of the very first imaging test a patient receives could streamline the entire pathway, potentially reducing unnecessary biopsies, guiding surgical planning toward nerve-preserving approaches for likely benign disease, and flagging suspicious lesions for earlier definitive workup. The authors position GobletNet as a tool for preoperative evaluation, and the reader study results suggest its most immediate value may be in supporting less experienced sonographers in settings where subspecialist head and neck radiologists are scarce.

As with any retrospective study, caveats remain. The model was trained and tested on data collected from five centers in China, and while the external validation sets demonstrate robustness across institutions, prospective deployment in real clinics, with live image acquisition and genuinely consecutive patients, is the next necessary step. The four-class framework also lumps heterogeneous entities into the other benign lesions category, and real-world performance on rare tumor subtypes remains to be established. Still, the combination of a large multicenter dataset, strong external validation, stable subgroup performance, and a demonstrated benefit to human readers places this work among the more convincing demonstrations of ultrasound-based AI in head and neck imaging. It also fits a broader pattern in which multi-task deep learning models, by learning to see and to interpret at the same time, are extracting diagnostic value from imaging modalities that have been in clinical use for decades but have never before been paired with algorithms this capable.

The study was supported by the National Natural Science Foundation of China and led by researchers at Zhejiang Cancer Hospital and the Hangzhou Institute of Medicine of the Chinese Academy of Sciences, along with collaborators at four other institutions. Published open access, the work arrives at a moment when health systems worldwide are searching for ways to expand specialist-level diagnostic capacity without proportionally expanding specialist headcount. If prospective studies confirm what this retrospective analysis shows, the humble ultrasound machine, already the most distributed imaging device on the planet, may gain a quiet digital colleague in the reading room, one that has studied more than seventeen hundred parotid tumors and rarely blinks.

Subject of Research: A multi-task deep learning model for ultrasound-based segmentation and four-class classification of parotid gland tumors

Article Title: An ultrasound-based end-to-end multi-task deep learning model to segment and classify parotid gland tumors: A retrospective multi-center study

Article References: Jiang, T., Wei, W., Cai, Y., Wang, K., Chen, C., Li, Q., Li, B., Song, M., Feng, N., Zhu, X., Pan, Q., Wang, H., Sui, L., Zhou, L., & Xu, D. (2026). An ultrasound-based end-to-end multi-task deep learning model to segment and classify parotid gland tumors: A retrospective multi-center study. BMC Medicine. https://doi.org/10.1186/s12916-026-05215-x

Image Credits: AI Generated

DOI: 10.1186/s12916-026-05215-x

Keywords: deep learning, parotid gland tumors, ultrasound, medical imaging, artificial intelligence, segmentation, tumor classification, pleomorphic adenoma, Warthin tumor, radiology, multi-center study, BMC Medicine