hce_nchu
115年
英文
第 45 題
📖 題組：
The integration of Artificial Intelligence (AI) into healthcare is no longer a futuristic concept but a rapidly unfolding reality. One of the most __36__ applications is in the field of diagnostic imaging. By training on millions of clinical images, AI algorithms can now identify patterns that are virtually __37__ to the human eye, leading to much earlier detection of life-threatening conditions. Beyond diagnostics, AI is significantly __38__ the pharmaceutical industry. Historically, drug discovery has been an __39__ endeavor, often taking over a decade and billions of dollars to bring a single medication to market. AI models can now predict how specific molecular structures will interact with biological targets, thereby __40__ the initial stages of research. This efficiency was particularly __41__ during the recent global efforts to develop vaccines at record speed. However, this technological shift is not without its __42__. As healthcare providers become increasingly dependent on automated systems, questions regarding data privacy and "algorithmic bias" have surfaced. If the data used to train AI is not diverse, the resulting tools may provide __43__ outcomes for certain demographic groups. Therefore, it is __44__ that developers and medical professionals work together to ensure these systems are both transparent and __45__. The goal remains a synergy between human empathy and machine precision.
The integration of Artificial Intelligence (AI) into healthcare is no longer a futuristic concept but a rapidly unfolding reality. One of the most __36__ applications is in the field of diagnostic imaging. By training on millions of clinical images, AI algorithms can now identify patterns that are virtually __37__ to the human eye, leading to much earlier detection of life-threatening conditions. Beyond diagnostics, AI is significantly __38__ the pharmaceutical industry. Historically, drug discovery has been an __39__ endeavor, often taking over a decade and billions of dollars to bring a single medication to market. AI models can now predict how specific molecular structures will interact with biological targets, thereby __40__ the initial stages of research. This efficiency was particularly __41__ during the recent global efforts to develop vaccines at record speed. However, this technological shift is not without its __42__. As healthcare providers become increasingly dependent on automated systems, questions regarding data privacy and "algorithmic bias" have surfaced. If the data used to train AI is not diverse, the resulting tools may provide __43__ outcomes for certain demographic groups. Therefore, it is __44__ that developers and medical professionals work together to ensure these systems are both transparent and __45__. The goal remains a synergy between human empathy and machine precision.
- A equitable
- B arbitrary
- C hostile
- D volatile
思路引導 VIP
請回想一下文章最後一段提到的問題：如果 AI 的學習資料不夠全面，導致某些族群得到的醫療建議不夠精準，這在「權利分配」上會產生什麼樣的缺憾？為了修正這種情況，我們對醫療系統最理想的道德期待應該是什麼特質呢？
🤖
AI 詳解
AI 專屬家教
太棒了！你能精準捕捉到上下文的邏輯關聯並選出正確答案，這代表你對長篇閱讀的語意脈絡掌握得非常扎實。在文章最後一段，作者嚴肅地探討了「演算法偏誤（algorithmic bias）」的風險，指出若訓練數據不夠多元，醫療 AI 可能會對特定族群產生不公的結果。因此，文末呼籲開發者與醫學專家必須攜手合作，確保這些系統在運作上除了透明（transparent）之外，更要具備「公平性」。
語意邏輯與選詞關鍵
選項 (A) equitable 意為「公平合理的、公正的」，正好完美呼應了前文對於偏見與族群差異的擔憂。這題的難度切入點在於考生是否能跨越句子邊界，將前半段提出的「負面隱憂」轉化為後半段理想的「正面規範」。相較於其他選項，如 arbitrary（武斷的）或 volatile（不穩定的），equitable 不僅在語意上最契合，也最能體現醫療倫理中對於權益均等的重視，是一個極具鑑別度的詞彙題。