# Artificial intelligence in psychiatric diagnosis: limitations and possibilities - URL: https://alisanburak.com/en/wcp-2026-posterler/ev0473 - Language: en - Last updated: not specified - Page title: Artificial intelligence in psychiatric diagnosis: limitations and possibilities | WCP 2026 Summary > Artificial intelligence can be a powerful diagnostic aid in psychiatry, but it cannot replace the psychopathologist; meaning, context, relationships, and judgm… This page was translated automatically from Turkish. [Read the original](https://alisanburak.com/wcp-2026-posterler/ev0473) Bu sayfa, 26. Dünya Psikiyatri Kongresi'nde sunulan çalışmanın yapay zekâ desteğiyle hazırlanmış Türkçe özetidir. Çalışmanın sahibi sunan yazarlardır; atıf için orijinal kaynağa başvurun. EV0473 E-PosterDijital Psikiyatri ve Yapay ZekâAnlatısal derleme ve uzman görüşü # Artificial intelligence in psychiatric diagnosis: limitations and possibilities ARTIFICIAL INTELLIGENCE IN PSYCHIATRIC DIAGNOSIS: LIMITS AND PROSPECTS Sunan: Marco O. Bertelli Ülke: Italy Kurum: CREA, San Sebastian Foundation, Misericordia di Firenze Artificial intelligence can be a powerful diagnostic aid in psychiatry, but it cannot replace the psychopathologist; meaning, context, relationships, and judgment remain the clinician’s responsibility. The poster aims to critically evaluate artificial intelligence in psychiatric diagnosis from psychopathological and phenomenological perspectives, considering both its diagnostic potential and the structural limitations of pattern recognition. The methodology consists of a narrative review and expert opinion integrating current evidence on AI-assisted psychiatric diagnosis with psychopathological, phenomenological, and hermeneutic perspectives; the approaches examined include classical machine learning (logistic regression, SVM, random forest), deep learning (CNN, LSTM, DNN), transformer and natural language processing (NLP) methods, including BERT, and multimodal models. The example studies presented in the poster report, in schizophrenia, 94.44% accuracy with DNN using multimodal MRI (n=144), 85% pooled accuracy and 81% leave-one-site-out accuracy with DANS using multicenter fMRI, 98% accuracy and an F1 score of 98.56% with RobIn using clinical features, and approximately 88% accuracy and an area under the curve (AUC) of up to 0.95 when combining NLP with clinician assessment of social media data. Speech-based AI assessment across eight common disorders achieved 93% accuracy, while a multimodal combination of NLP and fMRI achieved 85.7% accuracy; the poster notes that these examples should not be compared as benchmarks because their measurement methods, samples, and validation procedures differ substantially. The limitations listed include the inability to engage with subjective experience despite imitating empathic language, the interpretation of context-dependent and cultural expressions, intuition and timing in clinical interviews, the embodied relational context, and technical and ethical risks (data bias, poor external generalizability, black-box opacity, fabricated outputs, automation bias, privacy, and unclear accountability). The author argues that the central question is not whether AI improves prediction, but whether prediction is sufficient for psychopathological understanding. The recommendations include framing AI as a means of critical augmentation rather than replacement, maintaining human oversight and responsibility, using representative data and rigorous external validation, ensuring transparency and bias auditing, and preserving training in psychopathology and embodied clinical observation. [Orijinal postere git](https://cslide.ctimeetingtech.com/wcp26/attendee/eposter/file/2001) Bu çalışmaya nasıl atıf yapılır? APA 7: Bertelli, M. O., & Bianco, A. (2026, September 23-26). Artificial intelligence in psychiatric diagnosis: limits and prospects [E-poster]. 26th WPA World Congress of Psychiatry (WCP 2026), Stockholm, Sweden. https://cslide.ctimeetingtech.com/wcp26/attendee/eposter/file/2001 Vancouver: Bertelli MO, Bianco A. Artificial intelligence in psychiatric diagnosis: limits and prospects [e-poster]. 26th WPA World Congress of Psychiatry (WCP 2026); 2026 Sep 23-26; Stockholm, Sweden. Available from: https://cslide.ctimeetingtech.com/wcp26/attendee/eposter/file/2001 Yazar adları ve başlık biçimi kongre sisteminden otomatik oluşturulmuştur; kullanmadan önce orijinal posterle karşılaştırın. ## Aynı konudan - [Siber tanıkların davranışını belirleyen etkenler](https://alisanburak.com/en/wcp-2026-posterler/ev0041) - [Z kuşağında sosyal medyanın "lunapark etkisi"nin modellenmesi: örtük değişken yaklaşımı](https://alisanburak.com/en/wcp-2026-posterler/ev0046) - [DEHB'de dijital girişimler: etkililik ve çok modlu tedavi yaklaşımlarına entegrasyon](https://alisanburak.com/en/wcp-2026-posterler/ev0072) - [Algoritmaların inşa ettiği beyinler: dijital çağda DEHB, patolojiyi mi tanılıyoruz yoksa yeni bir normali mi patolojikleştiriyoruz?](https://alisanburak.com/en/wcp-2026-posterler/ev0077) - [Genç erişkin bir örneklemde nomofobi ile yaygın anksiyete bozukluğu arasındaki ilişki](https://alisanburak.com/en/wcp-2026-posterler/ev0111) [Tüm WCP 2026 özetleri](https://alisanburak.com/en/wcp-2026-posterler) --- Kaynak / Source: https://alisanburak.com/en/wcp-2026-posterler/ev0473 İletişim / Contact: Bağdat Caddesi, Veli Güneysu Apt. No:453 K:4 D:9, Suadiye / Kadıköy, İstanbul · +90 532 642 95 25 · info@alisanburak.com · https://alisanburak.com/bagdat-caddesi-psikoterapi-iletisim